Shoulder overuse injuries in the throwing athlete are a common clinical problem in sports medicine. From a physical assessment and treatment perspective, it is important for sports medicine clinicians to understand the long-term effects of throwing on the glenohumeral joint. Structural adaptations within the throwing shoulder of elite baseball pitchers have been widely reported, with much attention focused on the static stabilizing structures of the joint (1,5,6,14,15,24). Pitchers have been found to generate arm velocities >7000°·s−1 and rotational torques >70 N·m, with shear forces in the range of 300–400 N and compressive forces >1000 N (9,12). In addition, studies have shown that pitchers display a rotational range of motion (ROM) that is significantly greater than normal in their throwing shoulder (2,4,7,10,23,25,27). The extreme range of external rotation coupled with tremendously high joint forces during the throwing motion can exceed the physiological limits of the joint, thereby placing tremendous stresses on the static stabilizing structures (27). Over time, these combined stresses are thought to produce structural changes to the joint capsule and ligaments compromising joint stability (1,5,6,14,15,24).
Several theories exist that describe the pathoetiology of these soft tissue structural changes (1,5,6,14,15,24); however, quantitative evidence supporting these theories is lacking. One theory postulates that the anterior–inferior capsuloligamentous restraints “stretch out” or attenuate from the stresses of long-term throwing (1,14,15). The attenuated structures diminish joint stability by allowing the humeral head to sublux anteriorly during throwing. An opposing theory postulates that the posterior capsule gradually tightens with long-term throwing, predisposing the shoulder to injuries such as internal impingement and type II superior labral anterior to posterior (SLAP) lesions (1,5,6,24).
Passive glenohumeral stiffness is a reflection of the static structures resisting humeral head displacement from the cavitas glendoidalis (glenoid), and is characterized by the joint's ability to provide static stability and simultaneously control arthrokinematic behavior during upper extremity movement. Passive glenohumeral stiffness is quantified as the amount of force (N) required to displace a joint by a given amount (mm) (3,20). Passive glenohumeral stiffness measures, which provide information concerning the structural and mechanical properties of the joint, are considered to be clinically important when assessing joint stability (3,18,20). A stiffer joint is able to generate more force for a given displacement without disrupting the integrity of the capsuloligamentous restraints and, therefore, can decrease the risk of injury (3,19,29).
Passive glenohumeral stiffness can be assessed objectively using a force-displacement technique (2,3,17,18,20). An instrumented arthrometer can measure force-induced changes within the joint, enabling corresponding software to calculate the stiffness or passive resistance of the joint in response to applied forces (2,3). Instrumented arthrometers are becoming more readily available for the glenohumeral joint, with kinematic data beginning to emerge (2,3,11,26). Several arthrometric studies have found no significant differences in glenohumeral translation between the throwing and nonthrowing shoulders in baseball pitchers; however, data describing passive joint stiffness are lacking (2,11). Therefore, the present study was conducted to determine side-to-side differences in passive glenohumeral ROM and stiffness using selected kinematic measures in a group of asymptomatic professional baseball pitchers. We hypothesized that the throwing shoulder in asymptomatic pitchers would display greater passive glenohumeral ROM and stiffness than the contralateral nonthrowing shoulder. We based our hypothesis on the assertion that from long-term throwing and repetitive stresses to the joint the capsuloligamentous structures respond to the stress by improving the joint's ability to resist displacement during overhead throwing.
Design and Subjects
Instrumented shoulder stress examinations were performed on 34 professional baseball pitchers during the onset of spring training (mean age: 24.4 ± 3.7 yr; height: 188.3 ± 6.2 cm; body mass: 94.5 ± 9.6 kg). Stress examinations were performed along with selected glenohumeral ROM measures as part of the team's scheduled preseason physical examinations. Subjects reported pitching an average of 13.2 ± 6.5 yr. To be included in the study, both shoulders had to be asymptomatic at the time of testing and pitchers must have had no previous history of glenohumeral joint instability or shoulder surgery. All participants gave written informed consent, and an institutional review board approved the study protocol.
Instrumentation and Test Procedures
A standard plastic goniometer was used to measure passive glenohumeral ROM. Measurements of supine forward elevation (sagittal plane), internal and external rotation (90º abduction), and horizontal adduction (neutral rotation and maximal external rotation) were obtained using standard measurement guidelines (21). Two examiners performed all ROM measures. One examiner (MMR) held the shoulder position, while a second examiner (KEW) obtained the measurement after a firm endpoint was established. The supine position enabled the scapula to be stabilized, thus eliminating contribution from the scapulothoracic articulation during measurements. Examiners were blinded to the athlete's arm dominance.
Force-displacement measures were performed bilaterally using an instrumented stress device (LigMaster™, Sport Tech, Inc., Charlottesville, VA) attached to a mechanical frame for positioning the shoulder (TELOS Medical, Austin & Associates, Inc., Fallston, MD). The stress device contains a sliding pressure plate and linear encoder. A load cell embedded within the pressure plate measures forces applied to the joint in decanewtons (dN) and the linear encoder measures displacement in millimeters (mm). The TELOS frame has two adjustable counterbearings for stabilizing the shoulder girdle. The stress device is interfaced with a laptop computer where associated software is used for data reduction, storage, and display.
The LigMaster™ system uses a force-displacement model to analyze glenohumeral ligament function. Shoulders were positioned in abduction (90° in the scapular plane) and external rotation (60°) (Fig. 1). To limit scapular mobility during testing, the two counterbearings were placed against the coracoid process and scapular spine. Subjects were tested in standard fashion with the order for side (throwing or nonthrowing) and direction (anterior, posterior) counterbalanced. One force-displacement trial was performed for each direction for a total of four trials per subject. During each trial, a 15-dN displacement force was applied to the joint in a set direction (anterior or posterior) and rate (1 dN·s−1) with the stress device recording the force and displacement over the course of the trial. Each trial took approximately 15–20 s. The slow rate of force application enabled the stressed ligaments to respond in uniform fashion while keeping the surrounding musculature relaxed. Between each trial, the subject was instructed to remove his arm from the shoulder positioning frame and move it “back and forth” a few times immediately followed by repositioning the arm in the frame for the subsequent trial. This was done to “reset” or center the humeral head in the glenoid.
Force-displacement relationships were generated and displayed on screen (Fig. 2). For purposes of data analysis, the force-displacement curve is divided into two distinct regions (initial slope and final slope), which are delineated by approximating the curve as two linear sections. LigMaster™ uses a software algorithm to determine the best fit of the two lines by minimization of mean squared error. The slope (or stiffness) measures from the first region largely result from compression of the device padding coupled with soft tissue compression of the upper arm. As compression reaches its limits, further joint displacement then becomes constrained by the tensile force of the ligamentous structures being stressed. This defines the second region of the curve or the final slope. The initial slope and final slope characterize the passive resistance or stiffness of the joint to graded stress. The point at which the two regions intersect is referred to as the “inflection point” (shown as isolated points on the force-displacement graph). The inflection point is the point at which the final slope begins. The final slope value was used as the criterion measure for passive joint stiffness and was recorded in newtons per millimeter (Fig. 2).
Accuracy and repeatability of measurement.
Test repeatability and accuracy comprise an important part of the instrumented stress examination and are heavily dependent on patient (or subject) positioning (16). The clinical staff at Sport Tech, Inc. explicitly states in their user's manual that improper shoulder positioning can be a major source of error and strongly advise examiners to pay strict attention to detail and gain sufficient practical experience with the stress device apparatus and technique before testing patients. Studies performed by Sport Tech's own clinical staff have shown that test results in the hands of beginners are accurate to within 10%, but this improves to 4–5% after several trial runs. Ultimately, an examiner should be able to perform a stress study within 2% accuracy (16).
For the present study, our shoulder stress examiner (PAB) was adequately skilled and clinically proficient, having used the device in more than 100 trials before conducting this study. To ensure that our examiner had adequate technical skill, a pilot study was performed to determine repeatability and the precision of measurements using the LigMaster™ stress device. Glenohumeral joint stiffness was measured in 13 nonimpaired shoulders on two separate test sessions. The between session mean differences for stiffness were extremely small (<0.5 N·mm−1) with ICC values ranging from 0.29 to 0.89, depending on the side and direction (Table 1). The low ICC values (0.29 and 0.49) were specific to our anterior stiffness measures and are attributable to the small between subjects variability (8). Denegar and Ball suggest the inclusion of the standard error of measurement (SEM) values to demonstrate the precision of measurement (8). When comparing the SEM values with our mean stiffness values in Table 3, note that our measurements are extremely precise between test sessions.
Multiple paired t-tests were used to identify statistically significant differences in ROM between the throwing and nonthrowing shoulders. Because multiple (6) paired t-tests were used, a Bonferroni adjustment was performed a priori to control type I error rate with the significance level adjusted to P < 0.008.
A two within-factor ANOVA with repeated measures was used to identify statistically significant mean differences in passive joint stiffness between the throwing and nonthrowing shoulders for both anterior-directed and posterior-directed forces. In the presence of a significant interaction effect, pairwise comparisons were used to identify simple main effects. Level of statistical significance was set a priori at the 0.05 alpha level.
All data analyses were performed using SPSS® for Windows 11.0 (SPSS, Inc., Chicago, IL).
Summary data are displayed in Tables 2 and 3 for passive glenohumeral ROM and stiffness, respectively.
Paired t-tests revealed significant differences in passive ROM between the throwing and nonthrowing shoulder for external rotation and internal rotation at 90º of humeral abduction (Table 2). Side-to-side difference scores were significant for internal rotation (t(33) = 5.15, P < 0.008) and external rotation (t(33) = 3.51, P < 0.008). The throwing shoulder had significantly less (−8.5º) internal rotation than the nonthrowing shoulder. Conversely, the throwing shoulder had significantly more external rotation (5.1º) than the nonthrowing shoulder. The total arc of motion (internal + external rotation) was not significantly different between sides (t(33) = 1.65, P = 0.12) nor was forward elevation (t(33) = 0.812, P= 0.42) and horizontal adduction at neutral rotation (t(33)= 0.547, P = 0.59) and maximal external rotation (t(33) = 1.79, P= 0.08).
The side (throwing or nonthrowing) by direction (anterior or posterior) interaction effect was not significant (F(1,33)= 0.859; P = 0.361), nor was the main effect for side (F(1,33) = 0.154; P = 0.697). Bilateral comparisons show that passive joint stiffness was symmetric for both anterior-directed forces (TH: 16.6 ± 1.9 vs NTH: 16.2 ± 1.7 N·mm−1) and posterior-directed forces (TH: 15.1 ± 3.4 vs NTH: 15.3 ± 3.8 N·mm−1) (Table 3). The main effect for direction was significant (F(1,33) = 6.55; P = 0.015), with anterior joint stiffness (16.4 ± 1.6 N·mm−1) significantly greater than posterior joint stiffness (15.2 ± 3.2 N·mm−1).
Altered mobility patterns for internal and external rotation ROM have been consistently reported in the throwing shoulder of elite baseball pitchers (2,4,7,10,23,25,27). Adaptive change to the soft tissue structures of the glenohumeral joint is thought to be the cause. In the throwing shoulder, pitchers have been repeatedly shown to have an increased range of external rotation and a corresponding decreased range of internal rotation at 90º of abduction when compared with the contralateral shoulder (2,4,7,10,23,25,27). The magnitude of the gain in external rotation has been reported to be in the range of 8–12º, and this gain is usually offset by a symmetric loss of internal rotation. Interestingly, the total arc of rotation (internal + external) in the throwing shoulder has been consistently shown to be approximately 180 (± 15º) and is symmetric with respect to the contralateral shoulder (2,4,7,10,23,25,27). It appears that the total arc of motion in the throwing shoulder shifts “back,” favoring more external rotation. This ROM adaptation is thought to occur to allow greater arm cocking and, therefore, greater ball velocity on release (10). In the present study, external rotation was significantly greater and internal rotation was significantly less in the throwing shoulder compared with the nonthrowing shoulder; however, the total arc of rotation was symmetric. The gain in external rotation was found to be 5.1º and the loss in internal rotation was found to be 8.5º. Our findings are consistent with other studies showing a similar pattern of external rotation gain and internal rotation loss in the throwing shoulder, with the total arc of rotation being similar in magnitude to the contralateral shoulder (2,4,7,10,23,25,27).
Researchers have speculated whether structural adaptations compromise the static stability of the joint (1,5,6,14,15,24). In the present investigation, we used an instrumented stress examination technique to quantify the passive resistance or stiffness of the glenohumeral joint, thus providing insight into its static stability. During throwing, the glenohumeral ligament complex maintains joint stability by resisting humeral head displacement from the glenoid. We expected the throwing shoulder in professional baseball pitchers to display greater levels of passive stiffness when compared with the contralateral, nonthrowing shoulder. Furthermore, we expected that long-term stress to the dominant shoulder from repetitive throwing would result in soft tissue adaptations favoring greater passive joint stiffness. These expectations can only be applied to unimpaired shoulders with no history of instability or surgery.
To withstand the repeated stresses of pitching, greater than normal levels of passive stiffness would be beneficial for the pitcher by resisting joint displacement forces during pitching and, thus, acting as a prophylaxis against injury. In the present study, anterior-directed and posterior-directed forces were applied to the glenohumeral joint with the shoulder abducted and externally rotated to displace the humeral head in the respective direction. In this test position, anterior-directed forces primarily tested the structural integrity of the anterior-inferior capsuloligamentous restraints and posterior-directed forces tested the structural integrity of the posterior capsule. The bilateral comparisons that were performed in the present study found no significant differences in passive stiffness between the throwing and nonthrowing shoulder in our group of professional baseball pitchers. Based on our data, it appears that the resistive quality of the glenohumeral joint is not significantly improved with long-term throwing and, furthermore, is not compromised in light of the ROM adaptations that were found between the throwing and nonthrowing shoulders. Likewise, bilateral comparisons of glenohumeral translation in recent studies have found the throwing shoulder to have a similar amount of translational laxity when compared with the contralateral shoulder (2,7,11). Based on the results of these studies, it appears that static stability of the joint is not compromised, given the presence of a biomechanical shift in rotational ROM.
Other factors have been implicated in the development of this rotational ROM shift in the throwing shoulder of overhead athletes. Recent scientific evidence has attributed the rotational ROM shift, in part or whole, to structural changes of the proximal humeral growth plate of the throwing shoulder (7,23,25). Increased humeral retroversion in the throwing shoulder has been shown in several recent studies (7,23,25) and is thought to act as a controlling mechanism for the shift in internal and external rotation. In addition to providing a mechanism for increased external rotation, the osseous adaptation may also spare the glenohumeral joint from capsular disruption, thereby maintaining static stability. Support for this theory can be found from the results of the present study as well as the recent findings of Borsa et al. (2) and Ellenbecker et al. (11) showing positive rotational ROM shifts while also showing no significant side-to-side differences in glenohumeral translation between the throwing and nonthrowing shoulders of professional baseball pitchers.
When we compared passive glenohumeral stiffness between directions, posterior joint stiffness (15.2 ± 3.2 N·mm−1) was found to be significantly less than anterior joint stiffness (16.4 ± 1.6 N·mm−1). Structurally, the posterior capsule is thin compared with the anterior and inferior capsuloligamentous network (13,22,28). The relatively thin capsule may not be able to provide a comparable amount of resistance when posterior-directed forces are applied to the joint, leading to a tendency for greater posterior displacement of the humeral head during overhead activity. Borsa et al. (3) also noted a trend for noninjured shoulders to have less posterior than anterior stiffness. Less posterior stiffness may place the overhead athlete at a greater risk for glenohumeral joint injuries (e.g., instability, rotator cuff, and superior labral pathology). Injury prevention strategies could address this potential risk through strengthening regimens aimed at reinforcing the posterior cuff musculature.
This study examined only passive or static joint stiffness, whereas in normal shoulder function both passive and active restraining mechanisms interact to provide overall glenohumeral joint stability. To determine the overall stiffness of the glenohumeral joint, both active and passive joint restraints need to be assessed. Only professional baseball pitchers were used in this study; therefore, our findings can only be generalized to elite throwers.
The throwing shoulder of professional baseball pitchers was found to have significantly more external rotation and significantly less internal rotation than the contralateral shoulder. For passive joint stiffness, no significant differences were found between the throwing and contralateral shoulder. Overall posterior stiffness was significantly less than anterior stiffness. The repetitive stress of long-term throwing does create altered rotational patterns in the throwing shoulder of the professional baseball pitcher, but does not appear to compromise the joint's passive restraining quality.
1. Barber, F. A., C. D. Morgan, S. S. Burkhart, and C. M. Jobe. Current controversies. Point counterpoint. Labrum/biceps/cuff dysfunction in the throwing athlete. Arthroscopy
2. Borsa, P. A., J. Scibek, J. A. Jacobson, K. E. Wilk, M. M. Reinold, and J. R. Andrews. Glenohumeral laxity in professional baseball pitchers. Med. Sci. Sports Exerc.
3. Borsa, P. A., E. L. Sauers, and D. E. Herling. Glenohumeral stiffness response between men and women for anterior, posterior, and inferior translation. J. Athl. Train.
4. Brown, L. P., S. L. Niehues, A. Harrah, P. Yovorsky, and H. P. Hirshman. Upper extremity range of motion and isokinetic strength of the internal and external shoulder rotators in major league baseball players. Am. J. Sports Med.
5. Burkhart, S. S., C. D. Morgan, and W. B. Kibler. The disabled throwing shoulder: spectrum of pathology. Part I: Pathoanatomy and biomechanics. Arthroscopy
6. Burkhart, S. S., C. D. Morgan, and W. B. Kibler. Shoulder injuries in overhead athletes. The “dead arm” revisited. Clin. Sports Med.
7. Crockett, H. C., L. B. Gross, K. E. Wilk, et al. Osseous adaptation and range of motion at the glenohumeral joint in professional baseball pitchers. Am. J. Sports Med.
8. Denegar, C. R., and D. W. Ball. Assessing reliability and precision of measurement: an introduction to intraclass correlation and standard error of measurement. Journal of Sports Rehabilitation
9. Dillman, C. J., G. S. Fleisig, and J. R. Andrews. Biomechanics of pitching with emphasis upon shoulder kinematics. J. Orthop. Sports. Phys. Ther.
10. Ellenbecker, T. S., E. P. Roetert, D. S. Bailie, G. J. Davies, and S. W. Brown. Glenohumeral joint total rotation range of motion in elite tennis players and baseball pitchers. Med. Sci. Sports Exerc.
11. Ellenbecker, T. S., A. J. Mattalino, E. Elam, and R. Caplinger. Quantification of anterior translation of the humeral head in the throwing shoulder: manual assessment versus stress radiography. Am. J. Sports Med.
12. Fleisig, G. S., S. W. Barrentine, N. Zheng, R. F. Escamilla, and J. R. Andrews. Kinematic and kinetic comparison of baseball pitching among various levels of development. J. Biomech.
13. Gohlke, F., B. Essigkrug, and F. Schmitz. The pattern of the collagen fiber bundles of the capsule of the glenohumeral joint. J. Shoulder Elbow Surg.
14. Jobe, C. M., M. M. Pink, F. W. Jobe and B. Shaffer, Anterior shoulder instability, impingement, and rotator cuff tear: theories and concepts, In: Operative Techniques in Upper Extremity Sports Injuries,
F.W. Jobe (Ed.). St. Louis: Mosby, 1996, pp. 164–176.
15. Kvitne, R. S., and F. W. Jobe. The diagnosis and treatment of anterior instability in the throwing athlete. Clin. Orthop.
16. LigMaster's User Manual. Charlottesville, VA: Sport Tech, Inc., pp. 61–63.
17. Markolf, K. L., and H. C. Amstutz. The clinical relevance of instrumented testing for ACL insufficiency: experience with the UCLA clinical knee testing apparatus. Clin. Orthop.
18. Markolf, K. L., A. Graff-Radford, and H. C. Amstutz. In vivo knee stability: a quantitative assessment using an instrumented clinical testing apparatus. J. Bone Joint Surg.
19. McNair, P. J., G. A. Wood, and R. N. Marshall. Stiffness of the hamstring muscles and its relationship to function in anterior cruciate ligament deficient individuals. Clin. Biomech.
20. McQuade, K. J., I. Shelley, and J. Cvitkovic. Patterns of stiffness during clinical examination of the glenohumeral joint. Clin. Biomech.
21. Norkin C. C. and D. J. White, Measurement of Joint Motion: A Guide to Goniometry.
Philadelphia: FA Davis Company, 1985, pp 34–37.
22. O'Brien, S. J., R. S. Schwartz, R. Warren, and P. A. Torzilli. Capsular restraints to anterior-posterior motion of the abducted shoulder: A biomechanical study. J. Shoulder Elbow Surg.
23. Osbahr, D. C., D. L. Cannon, and K. P. Speer. Retroversion of the humerus in the throwing shoulder of college baseball pitchers. Am. J. Sports Med.
24. Pappas, A. M., R. M. Zawacki, and C. F. McCarthy. Rehabilitation of the pitching shoulder. Am. J. Sports Med.
25. Reagan, K. M., K. Meister, M. B. Horodyski, D. W. Werner, C. Caruthers, and K.Wilk. Humeral retroversion and its relationship to glenohumeral rotation in the shoulder of college baseball players. Am. J. Sports Med.
26. Tibone, J. E., M. G. Grossman, M. McGarry, S. Veneziana, D. J. Schneider, and T. Q. Lee. A cadaveric model of the throwing shoulder: a possible etiology of slap lesions. J. Shoulder Elbow Surg.
27. Wilk, K. E., K. Meister, and J. R. Andrews. Current concepts in the rehabilitation of the overhead throwing athlete. Am. J. Sports Med.
28. Wilk, K. E., C. A. Arrigo, and J. R. Andrews. Current concepts: the stabilizing structures of the glenohumeral joint. J. Orthop. Sports Phys. Ther.
29. Wright, V. Stiffness: a review of its measurement and physiologic importance. Physiotherapy
Keywords:©2006The American College of Sports Medicine
SHOULDER; STATIC STABILITY; INSTRUMENTED ARTHROMETER; EXTERNAL ROTATION; INTERNAL ROTATION