UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A comprehensive analysis of the swimmers' shoulder McKim, Kevin Robert 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-0066.pdf [ 5.65MB ]
Metadata
JSON: 831-1.0077314.json
JSON-LD: 831-1.0077314-ld.json
RDF/XML (Pretty): 831-1.0077314-rdf.xml
RDF/JSON: 831-1.0077314-rdf.json
Turtle: 831-1.0077314-turtle.txt
N-Triples: 831-1.0077314-rdf-ntriples.txt
Original Record: 831-1.0077314-source.json
Full Text
831-1.0077314-fulltext.txt
Citation
831-1.0077314.ris

Full Text

A COMPREHENSIVE ANALYSIS OF THE SWIMMERS' SHOULDER by KEVIN ROBERT M KIM C  B.H.K., The University of British Columbia, 1996  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE in  THE FACULTY OF GRADUTE STUDIES (School of Human Kinetics) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA December 1998  © Kevin Robert McKim,\<\W  In presenting  this thesis in partial fulfilment of the  requirements  for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for copying  of this thesis for scholarly purposes  department  or  by  his  or  her  representatives.  may be granted by the head of my It  is  understood  that  publication of this thesis for financial gain shall not be allowed without permission.  Department of  Sz/u*y(  4  <>$~  The University of British Columbia Vancouver, Canada Date  DE-6 (2/88)  M/A**^  extensive  J^^verrrrn  copying  or  my written  (  ABSTRACT Shoulder impingement syndrome is the most frequent injury suffered by the competitive swimmer. The literature is very thorough in describing the signs and symptoms of this condition. However, there is still a great deal of controversy in the published data about the etiology of shoulder impingement syndrome. Much of the debate undoubtedly exists because the cause of rotator cuff injury in one specific population may be different than in another population. For example elderly patients probably develop rotator cuff injury for different reasons than young athletes. The purpose of this study is to examine the role of seven common etiologic factors in the development of shoulder impingement syndrome. One hundred and seven competitive swimmers were recruited into the study. Each subject was classified as either acutely injured age-group (n=20), history of injury agegroup (n=35), non-injured age-group (n=40) or chronically injured master (n=12). The subjects were tested for differences in shoulder flexibility, glenohumeral joint laxity, posterior capsule inflexibility, rotator cuff strength and balance, rotator cuff endurance, scapulothoracic muscle strength, workload, swimming stroke technique, and acromial morphology. Workload and technique were evaluated with questionnaires. Shoulder flexibility and posterior capsule inflexibility were measured with a goniometer to determine range of motion. Joint laxity was tested by using anterior, posterior and inferior drawer tests. Rotator cuff strength and endurance were measured on a Cybex II Isokinetic Dynamometer. Finally, acromial morphology was examined in the injured subjects with a Supraspinatus Outlet View Radiograph.  The results seem to indicate that the shape of the acromion does not influence shoulder injury in a young athletic population. No significant differences were noted in the technique analysis among the age group swimmers. The test procedure was too insensitive to detect differences in the scapulothoracic muscles. The injured athletes had significantly reduced posterior capsuleflexibility.They were also found to have a significant degree of anterior and inferior glenohumeral translation. No differences were noted in the strength or muscle balance of the internal and external rotators of the shoulder. It appears that all competitive swimmers have strength imbalances between the strong internal rotators and the relatively weak external rotators. The injured subjects also had a trend towards reduced external rotation endurance. The mechanism of injury in the young competitive swimmer appears to be a combination of acquired joint laxity, rotator cuff imbalances and posterior capsule inflexibility. These mechanisms result in the anterior and superior migration of the humeral head. Humeral head migration causes compression and mechanical abrasion of the supraspinatus tendon between the head of the humerus and coracoacromial arch eventually leading to tendon inflammation, degeneration, and rupture. This study does not provide a definitive explanation for every case of shoulder impingement syndrome. It does provide relevant information about the condition of the competitive swimmer's shoulder and gives a possible explanation about the etiology of shoulder impingement syndrome.  iii  TABLE OF CONTENTS Abstract  ii  Table of Contents  iv  List of Tables  v  List of Figures  vii  Acknowledgements  viii  CHAPTER I  Literature Review  1  1.1 Rationale 1.2 Anatomy and Pathophysiology 1.3 Etiology of Shoulder Impingement Syndrome 1.4 Microvascular Anatomy of The Rotator Cuff 1.5 Acromial Morphology 1.6 Related Research  3 5 12 31 37 41  CHAPTER II  Hypothesis  50  CHAPTER III  Methods  55  3.1 Statistical Analysis  62  CHAPTER IV  Results  64  CHAPTER V  Discussion  84  5.1 Conclusion 5.2 Limitations and Future Research  102 105  Bibliography  107  Appendix A - Power Study  116  Appendix B - Consent Form and Questionnaire  118  iv  LIST O F TABLES TABLE  TITLE  PAGE  4 Incidence of Shoulder Injury/Pain in Competitive Swimmers Rotator Cuff Origin, Insertion, & Action 8 Jobe's Stages of Injury in Impingement of the Rotator Cuff 11 Results from Warner, Micheli, Arslanian Et Al. 1990 42 ROM Data from Beach, Whitney, and Dickoff-Hoffman, 1992 44 Strength Ratio Data From Beach, Whitney, and 44 Dickoff-Hoffman, 1992 44 Endurance Ratio Data from Beach, Whitney, and VII. Dickoff-Hoffman, 1992 45 Internal/External Rotator Ratio from Leroux, VIII: Codine, Thomas et al. 1994 Passive Range of Motion from Rupp, Berninger, and Hopf, 1995 46 IX. X. Rotation and ER:IR Ratio Results from Bak 49 and Magnusson, 1997 XI. Survey Information 57 XII. Subject Personal Data 64 65 Mean+S.D. Training Volume for Age Group Swimmers XIII. Competition Level of Age Group Swimmers 66 XIV. XV. Stroke Specialization for Age Group Swimmers 66 Mean±S.D. Age Group Swimming History and 66 XVI. Current Best Times 67 XVII. Current And Past Age-Group Shoulder Pain 68 XVIII. Results Of Age Group Impingement and Biceps Resistance Tests 69 Means+.S.D. for Age Group Flexibility Data XIX Means+S.D. for Age Group Joint Laxity 70 XX. 72 Mean+S.D. Age Group Rotator Cuff Internal XXI. and External Strength Data 73 XXII. Mean+S.D. Age Group Rotator Cuff Internal and External Endurance Data 73 XXIII. Mean+S.D. Stretching Data For Age Group Swimmers 74 XXIV. Mean+S.D. for Age Group Technique Analysis 74 XXV. Means+S.D. Training Volume Masters Athletes 75 XXVI. Means+S.D. Years Swimming, Race Distance, and Current Best Times 75 XXVII. Masters Current And Past Shoulder Pain XXVIII. Results Of Masters Impingement And Biceps Resistance Tests 76 76 XXIX. Means+.S.D. For Masters Flexibility Data 77 XXX. Means+S.D. For Masters Joint Laxity 77 XXXI. Mean+S.D. Masters Rotator Cuff Internal and External Strength Data I. II. III. IV. V. VI.  V  XXXII. XXXIII. XXXIV. XXXV. XXXVI.  Mean±S.D. Masters Rotator Cuff Internal and External Endurance Data Mean±S.D. Stretching Data for Masters Swimmers Mean+S.D. For Masters Technique Analysis Acromial Morphology Of Injured Age Group and Masters Swimmers Results Of Logistic Regression Analysis  78 78 78 79 83  vi  List of Figures Figure  Title  1. Graphic Representation of the Vascular Supply of the Humeral Head 2. Acromion Types Described by Bigliani and Morrison 3. Hypothetical Combination of Etiologic Factors in the Development of Shoulder Impingement Syndrome 4. Percentage Participation in Dryland Training Activities by Injury Category 5. Percentage Participation in Weight Training Activity by Injury Status 6. Percentage Stretching of Each Muscle Group by Injury Status 7. Total Swimming Related Injuries  Page  32 37 52 80 80 81 82  vii  ACKNOWLEDGMENTS I would like to thank Dr. Jack Taunton for his help and guidance through my entire masters program. I would also like to thank Drs. Robert Hawkins, Bruce Forster and Ron Mattison for their input into my research. A special thanks to Dr. Ted Rhodes and Rob Langill for their help and support in the collection of data for this study.  I. REVIEW OF LITERATURE Shoulder impingement syndrome is a common lesion in the athlete involved in repetitive overhead activities. It is the most frequent injury encountered in competitive swimming. Shoulder impingement syndrome is a chronic injury caused by repeated microtrauma of the subacromial bursa and supraspinatus tendon between the inferior aspect of the acromion, coracoacromial ligament, and greater tuberosity of the humerus. Several etiologic factors are thought to be responsible for the development of rotator cuff tendinitis: hypovascularity of the cuff, acromial variations, rotator cuff muscle imbalances, joint laxity, overwork, scapular dysfunction, and posterior capsule inflexibility. In the competitive swimmer, differences in joint laxity, posterior capsule inflexibility and rotator cuff muscle imbalances will be apparent between injured and uninjured athletes. Overwork also plays a significant role in the development of rotator cuff tendinitis in swimmers but there probably will not be a significant difference between groups. This study will investigate the significant factors associated with the development of shoulder impingement syndrome and determine their effects on swimmers. Testing will include rotator cuff muscle strength tests on a dynamometer, assessment of joint laxity and posterior capsule inflexibility, and questionnaires to determine workload. Additional testing will examine the incidence of scapular winging in both groups and assessment of acromial morphology in the injured swimmers. Measurement will be conducted on a group of swimmers with diagnosed shoulder impingement syndrome and on a group of healthy swimmers. The results of this study will demonstrate the significant etiologic factors in the development of shoulder impingement syndrome. This data will be used to develop a model to help identify and  l  diagnose athletes that may be susceptible to developing shoulder impingement syndrome. Team doctors, coaches, and physiotherapists could use the resulting information for the prevention and treatment of shoulder impingement syndrome.  2  RATIONALE Shoulder impingement syndrome is an extremely common problem in athletes participating in repetitive overhead activities. Impingement occurs frequently in athletes that participate in sports where the shoulder is abducted beyond 90° such as baseball pitching, tennis, volleyball, and swimming. In the 1960's, elite swimmers trained about 6000 meters per day. Currently, 21  21 81 87 89  most competitive swimmers cover between 8000 and 20 000 meters per day. ' ' ' Approximately eighty percent of practice time is spent performing the freestyle stroke.  11  Training seasons have also been expanded so that most athletes take at most one or two months rest per year. This workload translates to over 750 000-shoulder rotations per arm during an average season for elite swimmers 6 3  As a result, the shoulder is subject to strain by three mechanisms, namely a high repetition rate, the extremes of range used, and the force requirement of a sustained propulsive effort. 81  One effect of this workload is that rotator cuff dysfunction is the most common cause of shoulder pain in the swimmer. Incidence rates of between three and eighty 21  percent have been reported in swimmers, (see table I for reported rates) Richardson et. al. feel that incidence increases with the caliber of swimmer, both because training yardage will be higher and because they have been swimming longer. Shoulder 87  impingement most commonly develops in athletes competing in freestyle and butterfly strokes. ' 21  89  There are many theories that explain the development of shoulder impingement syndrome. Most of these theories have been developed by sports medicine physicians or orthopedic surgeons from their examinations of injured athletes. A few studies have used 3  cadavers to examine etiologic factors such as acromion morphology and shoulder vascularity. Even fewer studies have attempted to assess etiologic factors in athletes. Therefore, the purpose of this study is to examine the role of several common etiologic factors of shoulder impingement syndrome and to quantify their effects. The data from this study will be analyzed and a predictive model will be developed to assess the probability of an athlete developing shoulder impingement syndrome. The information gathered from this study can be used by coaches, sports medicine physicians, and physiotherapists in the prevention and rehabilitation of shoulder impingement syndrome. Author Kennedy & Hawkins Dominguez 87 Hall (unpublished) 87 Richardson et al.  Year 1974 1978  Ciullo21  1986  87  64 McMaster & Troupe  McKim & Guild (unpublished)  1980  1993  1996  Rate 3% 50% 67% 38% 23% 47% 57% 50% 68% 81.5% 46.7 38% 55% 64% 67% 75% 71% 55%  Sex both both male male female male female male female male female female male female male female male both  19% 9-35% Koehler 38-75% Table I. Incidence of shoulder injury/pain 55  1996  Level competitive high school College Non-elite Non-elite Elite Elite World Champion Team World Champion Team competitive adolescent competitive adolescent National Age Group National Age Group Senior Elite Development Senior Elite Development elite elite age group w/ history pain  age group w/ injury both both interfering shoulder pain history of shoulder pain both in competitive swimmers  4  ANATOMY AND PATHOPHYSIOLOGY The shoulder has adapted to become the most flexible and mobile joint in the human body. This great range of motion comes at a cost however; the shoulder is inherently unstable. Shoulder stability is maintained by static and dynamic stabilizers. The static stabilizers are the ligaments of the glenohumeral joint. Most of the stability in the shoulder is provided by the dynamic rotator cuff muscles. The rotator cuff muscles however, which are essential to maintaining glenohumeral integrity, are prone to injury and rupture. The shoulder joint is ball and socket synovial joint formed by the articulation of the humeral head with the glenoid fossa of the scapula. The joint is covered superiorly by the acromion and clavicle. The glenoid fossa is deepened by a ring of fibrous tissue, the glenoid labrum. "As the labrum expands the depth of the glenoid cup about the 81  humeral head, flexible stability and cushioning of bony impact are provided." Static stability of the glenohumeral joint is maintained by a series of intrinsic and extrinsic ligaments. The intrinsic ligaments reinforce the capsule, which surrounds the joint.  The capsule "is attached around the cartilaginous rim of the glenoid fossa and to  the anatomical neck of the humerus."  6 1  The capsule is reinforced by the glenohumeral  and coracohumeral ligaments. ' The coracohumeral ligament crosses superiorly over 61 108  the capsule passing from the base of the coracoid process to the tuberosities of the humerus. ' Bowen and Warren concluded that the coracohumeral ligament "was not a 61 108  significant ligamentous structure and did not contribute to the glenohumeral joint stability in a suspensory role." The glenohumeral ligaments "can be found anteriorly passing 1 6  from the anterior aspect of the glenoid fossa to the neck of the humerus.  6 1  There are three glenohumeral ligaments: superior glenohumeral ligament (SGHL), middle glenohumeral ligament (MGHL), and the inferior glenohumeral ligament complex (IGHLC).  16  "No single structure is primarily responsible for shoulder  stability at all positions of the arm." Therefore the ligaments must function together in 16  an attempt to maintain balance in the shoulder. The SGHL is the "primary capsuloligamentous restraint to inferior translation of the adducted, unloaded shoulder joint." It also has a secondary role supporting the "posterior capsule in limiting 16  posterior dislocation." (circle concept)  16  "The MGHL has the greatest variation in size and is absent more frequently than the other glenohumeral ligaments." "The contribution of the MGHL to static 16  glenohumeral joint stability is variable and most likely depends on its presence and size." Selective cutting has demonstrated that the "MGHL along with the subscapularis 16  tendon and superior part of the IGHL contributed to anterior stability at 45 degrees of abduction." It also has a role in limiting external rotation. 16  16  The IGHLC is "a complex portion of the capsule generally regarded as the structure most critical to glenohumeral joint stability." The fundamental function of the 16  IGHLC is to limit anterior-posterior translation. This ligament is also "the primary 16  stabilizer of the abducted shoulder." The IGHLC "has a primary but less significant role 96  as a inferior stabilizer at 0 degrees of abduction."  16  To accommodate the extremes of rotation that occur in multiple planes at the glenohumeral joint, the capsule must have some degree of looseness. At the same time, to act in the control of translation, there must be sufficient tension in the system. This fine balance is possible due to the anatomy of the capsule and the changing role of various components in resisting translation depending on the position of the arm. 16  6  A breakdown of this balance is often one of the principal ingredients in the development of shoulder injury. The intrinsic ligaments are essentially thickenings of the joint capsule. The extrinsic ligaments, on the other hand, are discrete structures separate from the capsule. One extrinsic ligament, the coracoacromial ligament and one tendon, the long head of the biceps, assists in the stabilization of the glenohumeral joint. The coracoacromial ligament extends from the coracoid process to the tip of the acromion. "It arches over the superior aspect of the joint, preventing upward dislocation of the humerus."  6 1  The tendon  of long head of biceps passes from the supraglenoid tubercle through the intertubercular groove to give rise to the muscle belly in the arm. "This tendon helps to stabilize the head of the humerus during movement." The glenoid labrum, joint capsule, ligaments 61  and some limited osseous structures function to provide static stability for the glenohumeral joint. The support provided by these stabilizers however is insufficient to maintain integrity of the glenohumeral joint. Therefore, adequate functioning of the dynamic stabilizers is of the utmost importance to ensure proper shoulder function. The rotator cuff muscles and their tendons provide dynamic stability of the glenohumeral joint. The cuff is composed of the supraspinatus, infraspinatus, teres minor, and subscapularis muscles, (see table II) Rotator cuff muscles serve three purposes in the shoulder: "contraction to create motion, contraction to restrain passive sources of motion, and relaxation when the force would be counterproductive."  81  The cuff appears to function by creating a stable fulcrum through which the deltoid can act to achieve arm elevation. If rotator cuff function becomes impaired for whatever  7  reason, upward displacement of the humeral head may occur and the normal fulcrum of the head on the glenoid is lost. '  14 16  The primary depressors during shoulder elevation are infraspinatus, teres minor, and subscapularis muscles. The supraspinatus functions "to maintain the humeral head 115  centered in the glenoid [and] contributes to arm elevation." The long head of biceps 104  also acts to depress the humeral head. Muscle  supraspinatus  Origin  Supraspinous,fossa  infraspinatus infraspinous fossa teres minor  upper 2/3 lateral border scapula subscapularis subscapular fossa  Insertion  Action  greater tuberosity, superior facet greater tuberosity, middle facet greater tuberosity, inferior facet lesser tuberosity  Shoulder abduction & external rotation Shoulder external rotation Shoulder external rotation Internal rotation  Table II. Rotator cuff origin, insertion, & action at 0° of shoulder abduction  Pectoralis major, latissimus dorsi, deltoid, trapezius, serratus anterior, and teres major are responsible for movement around the shoulder joint but provide little to stability of the glenohumeral joint. Shoulder impingement syndrome refers to the impingement of the soft tissues of the superior aspect of the shoulder (rotator cuff, subacromial bursa, and bicipital tendon) between the humeral head and the coracoacromial arch (coracoid process, acromion process, and coracoacromial ligament.) 26  There are two processes by which impingement may occur. Primary tendonopathy (intrinsic tendonopathy) is the gradual breakdown of the supraspinatus tendon resulting from age-related degeneration or calcification of the tendon. Secondary tendonopathy (extrinsic tendonopathy) results from any adverse change in the environment of the supraspinatus tendon. Subacromial crowding, posterior capsule tightness, or excessive superior migration may all play a role in the development of secondary tendonopathy.  Kamkar et al. and Allegrucci et al. defined "secondary impingement (secondary tendonopathy) as a relative decrease of the subacromial space due to instability of the glenohumeral joint or functional scapulothoracic instability." ' The degree of injury 2 52  caused by impingement was originally outlined by Neer in a three stage classification system. Stage I:  This stage is characterized by edema and hemorrhage within the  supraspinatus tendon and subacromial bursa. ' Typically, stage I lesions appear in 24 71  athletes under the age of twenty-five however, they can occur at any age. ' ' > > 19 24 52  74  71  These athletes will present with a positive impingement sign (forceful flexion of the shoulder pushes the head of humerus against the inferior acromion reproducing symptoms, pain is relieved by the injection of lOcc 1% Xylocaine or Lidocaine into the subacromial bursa.) The impingement test is useful for distinguishing these lesions from 19  a partiallyfrozenor subluxing shoulder. Common findings during examination include: 74  tenderness over the greater tuberosity, supraspinatus tendon, and anterior edge of acromion. ' Range of motion may be limited in some patients but a more common 19>24 52  sign is a painful arc of motion between 60° and 120° of abduction. ' ' Less common 19  24  52  symptoms include bicipital tendinitis, night pain, tenderness of the tendon within the bicipital groove, soft tissue crepitus, and painfully restricted range of motion. The key 24  differentiating characteristic between stage I and II impingement is that stage I is a reversible lesion that responds well to conservative treatment. '  52 71  Stage JJ:  Stage II lesions are a progression of stage I. "The physical findings in  stage II include those in stage I, with the distinguishing feature being that the impingement process is no longer reversible." During this stage, several changes occur 24  9  within the anatomy of the shoulder including bursal thickening, supraspinatus fibrosis and degeneration, and even partial thickness tears. ' Typical symptoms include soft 19 71  tissue crepitus due to scarring in the subacromial space, a catching sensation with lowering of the arm at approximately 100°, limitation of active and passive range of motion, and night pain. ' ' Stage II impingement is most common in individuals 19 24 52  between twenty-five and forty years of age. ' ' '  19 24 52 71  Stage HI:  The final stage of impingement is typified by partial or complete tears of  the rotator cuff, biceps tendon lesions, and bone changes at the acromion and greater tuberosity may be evident. ' Patients often present with limitation of shoulder range of 24 71  motion, which is more pronounced with active motion, infraspinatus atrophy, weakness of shoulder abduction and external rotation, acromioclavicular joint tenderness and biceps tendon involvement with rupture or degenerative changes. ' These individuals 19 52  will often also complain of extended periods of pain, especially at night. Complete tears and bony alterations usually develop in people over the age of 40 however cases have been reported in younger individuals. > > Surgery is always required to repair the 19  24,52  71  lesions associated with stage III impingement. Jobe has expanded Neer's original three-stage classification to a four-stage system. Stage I injuries are characterized by "inflammation with attendant swelling."  51  This often leads to muscle contraction and atrophy. Lesions of the cuff are described as stage II when it "can be assumed that there is some tissue fiber separation although the cuff tissue has not been completely divided." Stage I and II lesions are found 51  predominately in young athletes. . Treatment for stages I and II are usually conservative, consisting of rest, stretching, and strengthening exercises. In rare cases, surgery will be  10  required for stage U injuries to "decompress the coracoacromial arch in those who have healed with a thickened rotator cuff tendon."  51  If the injury progresses to include tearing of the rotator cuff, it is classified as stage IU or IV. Stage III injuries have tears of up to one centimeter while stage IV has lesions greater than one centimeter. Surgery is always required to repair stage n i and IV tears of the cuff. Stage  State of Fibers  I 77  Swelling, edema Dissociated  777 IV  Separated  Tear  None None  Treatment  Exercise Exercise, possible decompression < 1 cm Surgery > 1 cm surgery  Return to Professional Athletics  Yes Yes Qualified Doubtful  Table III. Jobe's stages of injury in impingement of the rotator cuff  11  ETIOLOGY OF SHOULDER IMPINGEMENT SYNDROME The signs and symptoms of shoulder impingement syndrome have been well described in the literature. A majority of authors describe similar clinical findings in athletes diagnosed with impingement. The area where discrepancy arises within the published literature is in the etiology of shoulder impingement syndrome. Many of the authors describe similar etiologic factors but emphasize one factor over another Overwork, joint laxity, posterior capsule inflexibility, hypovascularity, rotator cuff muscle imbalances and acromial variations are the most commonly cited causes of impingement syndrome. Some of the controversy in the literature maybe the result of the particular group being studied. The etiology of shoulder impingement syndrome may not be the same in swimmers as baseball players or young athletes and old. It appears that a complex interaction of factors is responsible for the development of rotator cuff tendinitis. No single factor is thought to be solely responsible for shoulder injury. This concept is apparent in a majority of articles on the subject. Over the years, new theories describing the etiology of shoulder impingement have been developed and old theories have been modified or discarded. This literature review comprises a chronological review of research into shoulder impingement syndrome. The first paper that described current surgical treatment and development of shoulder impingement syndrome, was presented by Neer in 1972. In addition to describing what at the time was a new surgical technique, anterior acromioplasty, Neer illustrated several factors related to the development of impingement syndrome. First, he points out that "the acromion revealed alterations attributable to mechanical  12  impingement." These changes include a "characteristic ridge of proliferative spurs and 70  excrescences on the undersurface of the anterior process [of the acromion], apparently caused by repeated impingement of the rotator cuff and humeral head. Neer also 70  described a "critical area for degenerative tendinitis and tendon rupture centered in the supraspinatus tendon, extending at times to include the anterior part of the infraspinatus tendon and long head of biceps." The significance of this area is discussed in later 70  papers. Neer's next paper on shoulder injuries in sport published in 1977 further describes the development of shoulder impingement syndrome. He states that the glenohumeral joint is prone to soft tissue injury because of its complex anatomy. "It has a great range 74  of motion made possible by the shallow glenoid and the stability provided by soft tissue."  74  Another unique feature of this joint is the subacromial space, which functions as a second joint cavity. It is predisposed to the derangements of repetitive loading, which result in the impingement lesions." 7 4  Neer also examined how the mechanics of shoulder motion could influence impingement. He found that the functional arc of elevation of the shoulder is forward, not lateral, and that the impingement occurs against the anterior edge of the acromion and coracoacromial ligament rather than against the lateral acromion. 74  Secondary impingement, Neer goes on to explain, may be caused by either thickening or separation of the acromioclavicular joint. Finally Neer presents some insight into the 74  development of shoulder impingement syndrome in swimmers. Swimmer's shoulder "may be found in one who breathes only on one side."  7 4  13  Swimmers were also the focus of study in paper presented by Kennedy, Hawkins, and Krissoff in 1978. Here the authors present the concept of mechanical impingement resulting from the demands of swimming. "In both the freestyle and butterfly strokes, the supraspinatus tendon and biceps tendon are called upon to do far more than nature originally intended." Both of these strokes require an athlete to internally rotate the 53  shoulder. Internal rotation tends to drive the greater tuberosity farther under the coracoacromial arch so that the impingement area is directly under the ligaments. '  53 87  The combination of forcing the 'critical zone' of the supraspinatus tendon under the coracoacromial arch and the repetitive nature of swimming results in habitual mechanical impingement. This chronic irritation leads to microtears and focal cell death in the avascular region of either supraspinatus or biceps tendon. Swimmers therefore, are 53  likely to develop impingement because the repetitive underwater pull and overarm recovery with subsequent mechanical impingement involved in strokes may invite a reactive tendinitis in a vulnerable area of the supraspinatus and less commonly biceps tendon.  53  Kennedy and Hawkins will emphasis in many future articles the importance of numerous etiologic factors in the development of shoulder impingement syndrome. The interaction of multiple components in the development of rotator cuff injury is further illustrated by Fowler's paper "Swimmer's Problem's" published in 1979. "There are a couple of anatomical and biomechanical problems inherent in this syndrome." The poor vascularization of the rotator cuff, specifically the supraspinatus 30  and biceps tendon, is the major consideration in the development of impingement syndrome. A result of freestyle and butterfly swimmers' training methods are 30  multiple episodes of'wringing out' the rotator cuff as the arm is brought down to the side and then abutted against the acromion as it comes around  14  and then abutted against the coracoacromial arch as the forward flexed shoulder is internally rotated. Athletes with 'tight shoulders' are also predisposed to impingement syndrome. Fowler lists the third key factor in the development of the swimmers' shoulder as faulty stroke mechanics. "The person with the higher arm recovery seems to have fewer problems."  30  Fowler also suggests correcting athletes that breathe consistently to one side to prevent injury from occurring. The most effective treatment for shoulder impingement is 30  prevention. "There is no question that exercises in the preseason prevent many of the problems to the shoulder." The preseason program should include flexibility and 30  strengthening exercises to prevent shoulder impingement syndrome before it occurs. Hawkins and Kennedy published a second paper on impingement syndrome in 1980. In this article they expanded on two etiologic factors mentioned in their original paper: overwork and avascularity. When the shoulder is abducted, the blood vessels of the supraspinatus tendon are completelyfilledbut when adducted an area of avascularity is present. This critical area is "one centimeter proximal to the point of insertion" of the 42  supraspinatus tendon. Degeneration of the tendon then may be a result of the constant 53  avascular pattern associated with shoulder rotations. Overwork also plays a role in 53  shoulder degeneration as the "mechanical impingementfromrepetitive loading" functions to breakdown the structural components of the shoulder.  53  There are two mechanisms that can lead to the development of shoulder impingement syndrome. These were described by Penny and Welsh in 1981. The first mechanism occurs when the volume of structures passing beneath the arch is increased. With less space available, there is a greater chance of impingement. In addition, the  15  vascularity of the shoulder is compromised when the arm is adducted to the side. This leads to a "wringing out" effect in the supraspinatus and biceps tendons.  80  Further impingement occurs if there is a partial tear of the supraspinatus muscle involving its inferior surface. This allows buckling of the tendon as it passes beneath the coracoacromial arch, resulting in a painful catching sensation. 80  Rotator cuff tendinitis can also result when the space available for the rotator cuff diminishes and impingement is more prone to occur. Reduction of available space is •  80  often the result of osteophytes on the inferior and anterior aspects of the acromion. "This reduces the amount of space available for the cuff during abduction and acts as a mechanical irritant exacerbating impingement." These mechanisms were later 80  described as primary and secondary shoulder impingement. Strength imbalances between the internal and external rotators of the muscular cuff also contribute to the development of impingement. In a study of the biomechanics involved in swimming, Richardson demonstrated that the "pull-through phase consists of adduction and internal rotation." It is this phase of the swimming stroke that selectively trains the internal rotators of the cuff muscles. Humeral head migration may occur if 89  dryland training does not address differences in muscle strength. The rotator cuff acts as a depressor of the humeral head, responsible for preventing if from migrating proximally during adduction. If the rotator cuff is overwhelmed by the deltoid, impingement of the rotator cuff is inevitable. 89  A second paper by Richardson and Miller in 1991 concluded that impingement in swimmers is affected by flexibility and swimming biomechanics. Biomechanical errors 88  are related to the amount of body roll used when swimming frontcrawl. "Less roll leads to an increased incidence of shoulder pain."  88  16  The development of shoulder injuries in swimmers was further examined by Fowler and Webster in 1983. They hypothesized that "coracoacromial arch impingement coupled with relative avascularity figure prominently in the etiology" of shoulder injuries in swimmers. Secondary factors include "stroke technique and the athlete's unique 31  anatomical configuration." Imbalances in the rotator cuff muscles were also enhanced 31  in swimmers because the internal rotators are used more frequently in freestyle and butterfly swimming strokes. Training distances were not related to the development of shoulder injury. Nor was there a relationship between shoulder pain and posterior 31  subluxation.  31  Shoulder impingement syndrome is probably not the result of one or even two isolated etiologic factors. The accumulated effect of several elements result in rotator cuff lesions. Hawkins and Abrams introduce this theory in their paper "Impingement Syndrome in the Absence of Rotator Cuff Tear" published in 1987. They begin by stating that "individuals who excessively use their arms above the horizontal are especially at risk for developing impingement." Upper extremity positions affects the 41  circulation within the cuff creating a zone of avascularity. Repetitive subacromial loading combined with the vulnerable avascular region result in tendinitis. "Subjecting 41  an area of tendinitis to repetitive irritation with compromised metabolism has been shown to alter the normal biologic consistency of the tissues." If this condition is not 4 1  remedied, "progressive wearing and attrition occurs within the tendons resulting in microtears and partial thickness tears of the rotator cuff. Contributing factors may be 41  abnormal shape or thickness of acromial process, incompletely fused apophysis, and prominence of the greater tuberosity of the humerus. Arch variations can also make a 41  17  shoulder more susceptible to impingement. Hawkins and Abrams conclude by stating 41  that degenerative cuff tears, normally seen only in older individuals, are seen in younger athletes because of "extremes in aggravating activities."  41  Butters and Rockwood in 1988 also describe an interaction of events in the development of impingement. First, impingement usually is secondary to an offending prominent anterior acromion. Variation in acromial shape can be correlated with the occurrence of impingement and degeneration of the rotator cuff. 19  A type III hooked acromion is more frequently associated with impingement syndrome. The "pathophysiology of the impingement syndrome includes vascular and mechanical explanations that are complementary." Early studies have identified a 19  hypovascular zone in the supraspinatus and biceps tendon. The pattern of "relative avascularity in the supraspinatus tendon near its insertion as well as bicep tendon" contribute to the development of rotator cuff tendinitis. The second factor related to 19  development of tendinitis is mechanical impingement of the supraspinatus on the anterior acromion. Mechanical impingement occurs against a prominent anterior acromion causing "degenerative changes occurring in the supraspinatus tendon"  19  Fowler's paper published in 1990 presents a theory of shoulder impingement that is incorporated into the model for this study. Once again, the interaction of several elements are cited as the cause of shoulder impingement. The association of "anatomical features and biomechanical forces combine to produce 'swimmer's shoulder.'" The 32  first factor is overwork. The shoulder is "most vulnerable to injury in the overhead position. Swimming puts continuous repeated demands on the shoulder in this position." The rotator cuff muscle must work hard to minimize movement of the 32  18  humeral head which eventually leads to fatigue. "Superior migration of the humeral head may occur with cuff fatigue, increasing subacromial loading." A second cause is the 32  mechanical impingement associated with swimming. Repeatedly the tendons are impinged against the arch, which may result in a mechanical irritation and an inflammatory response or tendinitis. '  32 113  Acromial variations and hypovascularity also  play a role in the development of impingement. ' Increased shoulder laxity may also 32 113  be responsible for the evolution of tendinitis. Lax shoulders place "additional demands on already fatigued muscles." and "there does seem to be a relationship between tendinitis and laxity." A final component is an imbalance in shoulder strength. A study 32  by Fowler found that "swimmers have an abnormal ratio of internal to external rotators." As noted earlier, swimming selectively trains the internal rotators. If strength 32  imbalances are not corrected, abnormal migration of the humerus may occur. In conclusion, Fowler states that 'overwork, impingement, and hypovascularity are three main factors the contribute to impingement tendinitis in the competitive swimmer."  32  Abrams' paper from 1991 states that "impingement syndrome is a secondary process." The primary process is humeral head subluxation and protraction of the 1  scapula which reduces the subacromial space. "Rotator cuff degeneration is caused by 1  exterior wear on the undersurface of the subacromial arch." Two mechanisms exist by 1  which impingement occurs. First, thickening of the bursae and tendons further narrow the space separating the supraspinatus from the undersurface of the acromion, acromioclavicular joint, and coracoacromial ligament. Chronic abrasions create 1  hypertrophy of the subacromial space that leads to spur formation and thickening of the  19  ligaments. The alterations within the subacromial space further increase mechanical 1  impingement. The second mechanism of impingement results from "occult instability" of the shoulder. Dynamic stability of the glenohumeral joint is provided by the rotator cuff and 1  long head of the biceps. Compromise of the rotator cuff permits upward translation of the humeral head, narrowing the subacromial space. Laxity of the joint also permits 1  abnormal head translation. Laxity of the shoulder must be compensated with increased 1  rotator cuff activity. As "the cuff fatigues, tensile injury causes further dysfunction."  1  Superior migration of the head compromises the subacromial space, leading to cuff failure, secondary impingement and changes within the rotator cuff and subacromial bursae.  1  The anatomical structures that prevent abnormal shoulder movement were thoroughly investigated by Bowen and Warren in 1991. "The anatomy and geometry of the glenohumeral articulation provide little inherent stability." The dynamic and static 16  constraints are therefore crucial in maintaining stability in the shoulder. The bony and capsuloligamentous structures comprise the static constraints.  16  The muscle tendon units of the rotator cuff muscles may have some static role in limiting excessive translation of the glenohumeral joint. Their main function appears to be a dynamic one. 16  The dynamic stabilizers function by compressing the humeral head into the glenoid. If 16  the dynamic and/or static glenohumeral stabilizers begin to fail, injury to the rotator cuff is likely. The authors believe that injury to the rotator cuff, such as that which occurs due to overload in an overhead athlete, may lead to diminished joint compression. As a result, increased translation in the anterior-posterior or superior-inferior direction may be possible. This excessive translation may cause overload of  20  the capsuloligamentous structures and stretching or failure of them. The increased translation may also cause shear forces on the glenoid and result in labral injuries. 16  The process by which damage to the static and dynamic stabilizers occurs was examined by Pollock et al. They studied the mechanical properties of the inferior glenohumeral ligament complex (IGHLC) and found that it undergoes considerable plastic deformation or stretching before ultimate failure. This observation helps to explain how repetitive microtrauma can cause gradual stretching of the ligaments and clinical subluxation. 83  Many articles on shoulder impingement syndrome have suggested that shoulder instability is a primary cause of rotator cuff injury. The research by Bowen and Warren and Pollock et al. provides an explanation as to how the repetitive nature of competitive swimming eventually leads to joint laxity and injury. The datafromBowen and Warren's study also provide support for the "circle concept" of shoulder stability. For example, 16  Warner at al. concluded that the physical finding of a sulcus sign at 0 degrees of abduction reflects laxity of the superior structures (SGHL) and negative intra-articular pressure in the shoulder joint. 16  A positive anterior-posterior drawer would reflect laxity of the inferior glenohumeral ligament. In 1991 Chansky and Iannotti published a review paper addressing the vascularity of the rotator cuff. The main supply to the humeral head and tendons of the rotator cuff is through the arcuate artery, a branch of the anterior circumflex artery. The 20  suprascapular artery delivers blood to the supraspinatus muscle and tendon. This article 20  is one of the first to draw attention to the controversy regarding vascularity of the cuff. They quote several cadaver studies that report a critical portion in the supraspinatus  21  tendon that appears anemic and susceptible to degeneration. Contradictory to these 20  findings, they cite Swiontkowski et al. who found that "impingement induces a hyperemic response that results in resorption of previously damaged collagen fibers."  20  Changes in the subacromial space is listed as key factor in the development of rotator cuff lesions in Scarpinato et al. (1991) paper. They explained that primary compressive rotator cuff disease is "associated with type III hooked acromion, degenerative spurs, Os Acromiale, or congenitally thick coracoacromial ligament" which reduce the subacromial space. Impingement may also be secondary to another 95  underlying problem such as glenohumeral instability. Rotator cuff failure may also 95  cause secondary impingement. Repetitive tensile overloading of the cuff exceed the ability of both dynamic and static stabilizers to compensate leading to humeral head translation and secondary impingement."  9 5  Silliman and Hawkin's paper published in 1991 presents several new ideas in the development of shoulder impingement syndrome. Their hypothesis is the "circle concept of instability." According to this theory, excessive translation in one direction probably 97  is related to excessive translation in other directions. ' 9 7  1 U  Subtle and varying degrees of  instability are responsible for pain and dysfunction in the athlete's shoulder. '  60 97  Instability, in this theory, is the prime factor in the development of impingement. Secondary factors include mechanical abrasions and a "hypovascular water shed area in the critical zone of the supraspinatus tendon." Additionally, excessive force acting on 97  the rotator cuff associated with repetitive overhead actions often times exceed the physiologic limits of the tendon's ability to sustain those forces. Finally, Silliman and 97  22  Hawkins outline the debate over the role of acromial morphology in the development of impingement syndrome. They state that present investigations are underway to determine whether acromial morphology is a cause of rotator cuff disease or more a result or reaction to chronic rotator cuff degeneration. 97  Silliman and Hawkins expanded on a classification system to evaluate athletes with shoulder pain designed by Jobe et al. The four point system is: Group I Group II  athletes who demonstrate pure impingement athletes who have instability secondary to anterior ligament and labral injury with secondary impingement athletes who have instability due to hyperelasticity and secondary impingement  group UI group IV  athletes who have pure anterior instability  97  "This classification system suggests a significant overlap between rotator cuff pathology and instability of the shoulder girdle in overhead athletes."  9 7  Kamkar et al. in 1993 reiterate several key etiologic factors and suggest a couple of new ideas in the development of shoulder impingement syndrome. Factors that minimize impingement include: adequate shape of the coracoacromial arch; normal inferior surface of the acromion; normal function of the humeral head depressor mechanism, and normal capsular laxity. There are several factors that lead to the development of impingement. First, 52  fatigue of the serratus anterior muscle from repetitive use may cause impingement because of improper temporal sequencing of scapulothoracic musculature." 52  Mechanical impingement of the rotator cuff underneath the coracoacromial arch, termed primary impingement, results from: subacromial crowding, posterior capsule tightness, and excessive superior migration of the humeral head due to weakness of the humeral head depressors. 52  23  "Secondary impingement is defined as a relative decrease of subacromial space due to instability of glenohumeral joint or functional scapulothoracic instability."  52  Bony changes and spur formations have also been reported as causes of impingement. "Bony changes are caused by repeated impingement of the rotator cuff between the coracoacromial ligament and the humeral head." Tightness of the posterior 52  capsule is proposed to contribute to impingement syndrome causing superior translation of the humeral head during shoulder flexion and cross-chest adduction. Finally, fatigue 52  of the rotator cuff can cause abnormal translation of the humeral head. Impingement 52  may be aggravated by weakness of the scapulothoracic muscles. Regardless of the 52  cause, the end result is rotator cuff tendon inflammation and potential rupture.  52  Proper functioning of the rotator cuff is once again emphasized in a paper by Meister and Andrews. They claim that "overhand athletes are dependent on the rotator cuff due to acquired laxity of the glenohumeral joint." Others factors besides laxity 66  contribute to the development of impingement. These include; morphology of the acromion, congenitally thick coracoacromial ligament, prominent coracoid lip, degenerative spurring of the acromion, and posterior capsule tightness. In swimmers, 66  two factors in particular play a key role in impingement. These are hyperelasticity in the joints of swimmers and muscle imbalances due to "sustained internal rotation" during swimming.  66  Miniaci and Fowler in 1993 report "an association between rotator cuff tendinitis and shoulder laxity in competitive swimmers." Shoulder laxity and imbalances of 67  muscles about the shoulder can cause repeated episodes of subluxation, leading to inflammation. Swimmers develop muscle imbalances by selectively training the 67  24  internal rotators to improve power and speed therefore, imbalance between internal and external rotation is a contributing factor. Other contributing factors include a type III 67  acromion morphology, avascularity, and muscle weakness or fatigue. Finally, scapular 67  winging is a consistent and reliable finding suggesting underlying shoulder pathology is associated with scapulothoracic dysfunction.  67  The repetitive nature of swimming strokes plays a fundamental role in the development of shoulder impingement syndrome according to Allegrucci, Whitney, and Irrgang. Muscular fatigue of the rotator cuff and scapula stabilizers may develop after prolonged use, leading to increased anterior and superior humeral head translation and eventually secondary impingement and resultant rotator cuff tears. 2  Additionally, swimmers have "been identified as having increased shoulder laxity as a consequence of the physical demands placed on the shoulder during athletic activity."  2  The authors believe that constant overuse associated with intense training results in joint laxity and eventually leads to impingement. The combination of muscle fatigue and joint 2  laxity make the swimmers shoulder very susceptible to overuse injuries. Other predisposing factors include weakness or muscle imbalances that "disrupt the normal scapulothoracic rhythm", improper swimming biomechanics, and unilateral breathing.  2  Bowen and Warren (1994) report overuse, collagen failure and progressive attrition of tendon as possible etiologies in the development of rotator cuff tendinitis.  17  Other factors include decreased vascularity of the cuff, impaired rotator cuff function, microtrauma, fatigue, abnormal bony anatomy, and instability.  17  Corso reported in 1995 that  25  muscle weakness, tightness of the inert tissue that precludes accessory motion, and poor scapulohumeral rhythm can result in altered biomechanics, shoulder dysfunction, and impingement syndrome. 24  The function of the rotator cuff is to provide joint stabilization, approximation, and centralization of the humeral head in the glenoid fossa." Fatigue of the rotator cuff, 24  24 28  deltoid, and/or long head of biceps leads to abnormal translation of the humerus. ' "Weakness results in cephalad migration of the humeral head due to loss of depressors. Superior migration increases impingement and reinforces the cycle." Shape and slope of 28  24 28  the acromion also plays a significant role in the development rotator cuff injury. ' There is an increased incidence of rotator cuff tears in persons with type II and III acromion. Avascularity of the region, and chronic irritation also play a significant role 28  in the development of impingement syndrome. '  24 28  In 1996 Koehler and Thorson describe the development of shoulder impingement in competitive swimmers. This lesion is most common in freestyle and butterfly swimmers. The shoulder is subject to impingement when it is in the early to mid-pull 55  through phase, which involves extreme adduction and internal rotation. Rotator cuff 55  fatigue, scapular dysfunction, and multidirectional laxity all contribute to impingement.  55  Finally, the supraspinatus tendon is poorly vascularized and adduction and internal rotation at the end pull-through phase may contribute to tendinitis by compressing the tendons vascular supply. 55  Koehler reiterates the hypothesis that a combination of factors are responsible for the development of rotator cuff injury in the competitive swimmer. The development of rotator cuff injury in the general population appears to be "multifactorial and is still a subject of debate and investigation" according to Blevins.  14  26  He proposes that injury is the result of intrinsic and extrinsic factors. Intrinsic factors include hypovascularity and primary age related degeneration while impingement, repetitive microtrauma and macrotrauma are classified as extrinsic factors. The 14  etiology of shoulder impingement syndrome in athletes differs significantly from people in the general population. Although primary impingement, vascular supply, and primary degenerative changes may contribute, trauma appears to play a greater role. Repetitive microtrauma glenohumeral instability and macrotrauma appear to be significant factors in the development of cuff pathology in athletes. 14  Three mechanisms of rotator cuff injury are apparent in the overhand athlete. "The cuff maybe injured as a result of repetitive microtrauma (as seen in the overhand athlete with primary impingement), primary tensile overload, or impingement and tensile overload secondary to instability."  14  Primary impingement is usually related to abnormal acromial architecture. "A hooked acromion or os acromiale may limit the available subacromial space" thereby increasing the amount of mechanical abrasion on the tendons. This process is often 14  associated with the presence of osteophytes or bone spurs on the inferior surface of the acromion and/or the acromioclavicular joint. Primary impingement is most commonly 14  seen in overhead athletes over the age of forty.  14  Primary tensile overload occurs more frequently in throwing athletes but the process may have an influence in swimmers. Primary tensile overload occurs when the rotator cuff is stressed beyond its ability to adapt and heal resulting in tendon collagen failure, inflammation, and degeneration. ' This typically occurs with "a sudden 14 18  increase in the intensity or duration of the throwing activity beyond a level to which the  27  athlete is accustomed" A similar process may occur in swimmers that rapidly increase 14  their workload without allowing adequate time for recovery and adaptation. The third process described by Blevins is "subtle anterior glenohumeral instability resulting in secondary mechanical impingement and tensile overload." The static 14  stabilizers (primarily the anterior joint capsule and inferior glenohumeral ligaments) are placed under "considerable stress in the overhead athlete."  14  The static stabilizing function of the glenohumeral ligaments may be compromised by inherent laxity, gradual stretching from repetitive stress or macroscopic failure. Excessive translation due to failure of static capsuloligamentous constraints may cause cuff injury by direct mechanical compression (secondary impingement) or through the increased loads on the cuff that occur during distraction or shear of the humeral head on the glenoid (secondary tensile overload). Increased stress on the cuff is produced as it works to limit excessive translation to maintain stability. 14  The presence of a tight posterior capsule may also influence the development of secondary impingement by increasing "anterior shear forces." A tight posterior capsule 14  has been shown to "increase anterior translation of the abducted flexed arm."  1 4  Physical examination of athletes with shoulder impingement may provide insight into the nature of their injury. "Mild to moderate winging of the scapula is most commonly associated with scapular stabilizer muscle asynchrony and dysfunction resulting from chronic shoulder pain." Increased anterior translation on laxity testing is 14  a frequent finding in "athletes with rotator cuff pathology secondary to subtle anterior instability." Finally, a sulcus sign greater than two centimeters may be associated with 14  "multidirectional instability which can contribute to impingement and cuff pathology."  14  The current theories on the etiology of shoulder impingement syndrome have been thoroughly discussed in papers by Hulstyn and Fadale (1997) and Wolin and Tarbet (1997). Several potential mechanisms are discussed. The first two have been reported  28  previously in the literature. Primary impingement occurs when repetitive overhead activity results in impingement of the supraspinatus against the anterior inferior aspect of the acromion and/or the coracoacromial ligament. ' Secondary impingement is caused 47 116  by instability of the glenohumeral joint. Repetitive microtrauma or macrotrauma place increased demands on the rotator cuff as it attempts to keep the humeral head centred in the glenoid.  47,116  The constant stress on the glenohumeral static stabilizers eventually  results in instability of the joint. "Progression of the instability can lead to subluxation 6  and eventual secondary impingement of the rotator cuff. Fatigue, intrinsic injury, and 6  partial undersurface tears may ensue.  116  A third mechanism of injury is tensile failure resultingfromthe eccentric deceleration associated with throwing sports. ' -  6 47 116  Repetitive cycles of large forces generated and then dissipated during overhead activities place stress on the soft tissues of the shoulder girdle. The initial response to excessive stress on the rotator cuff and surrounding tissues is reversible inflammatory changes. With continued overuse, intrasubstance microtraumatic injury causes structural changes that act as a stress raiser. 47  Progressive injury occurs, eventually leading to rotator cuff dysfunction and rupture. This process maybe exacerbated by "a tenuous blood supply at the rotator cuff insertion."  47  Thefinalmechanism is termed internal posterior superior glenoid impingement or intra-articular cuff impingement. This process is thought to occur in conjunction with atraumatic shoulder instability. ' ' When the arm is overhead abducted and maximally 18 37 47  externally rotated the posterior inferior aspect of the supraspinatus is impinged between the greater tuberosity of the humeral head and the posterior superior labrum. ' < > ' > 6 14  18  37 47  116  This may end in an undersurface tear of the posterior aspect of the supraspinatus.  116  29  Different degenerative processes seem to occur in young and old athletes. Young athletes typically develop secondary impingement resulting from joint laxity. "Anterior 6  instability is frequently present and may be the primary pathology leading to a secondary subacromial impingement lesion." "Primary bony impingement can occur in young 6  throwers but this is rare." It is more commonly seen in those athletes with a prominent 6  type III acromial morphology.  6  The older athlete, on the other hand, often has "degenerative processes associated with mechanical impingement on the coracoacromial arch." Thickening of the 6  coracoacromial ligament and spur formation which compromise the subacromial space are common findings in the masters athlete.  6  30  MICROVASCULAR ANATOMY OF THE ROTATOR CUFF The role that rotator cuff vascularity plays in the development of shoulder impingement syndrome is currently open to debate. Many classic studies have found this area to be hypovascular and many authors have attributed this as a leading cause of rotator cuff lesions. However, recent studies have found the cuff to be well perfused with blood. A definite answer about the effect of rotator cuff blood supply on the development of impingement cannot be given at this time. Both views on the issue will be presented below. The muscles and bones around the glenohumeral joint receive blood from branches of the axillary and brachial arteries. Rothman and Parke in their study in 1964 found that six arteries regularly supply the muscles of the rotator cuff. The arteries and the percentage of cadavers they were discovered in are: 1. 2. 3. 4. 5. 6.  suprascapular anterior circumflex humeral posterior circumflex humeral thoracoacromial suprahumeral subscapular  100% 100% 100% 76% 59% 38%  9  "The posterior circumflex humeral and suprascapular arteries form an interlacing pattern on the posterior part of the cuff with several large anastomoses."  91  This is the  predominant artery to the teres minor and infraspinatus. "The anterior circumflex 91  humeral artery approaches the anterior aspect of the rotator cuff over the subscapularis muscle" then forms an anastomoses with the posterior circumflex artery in the region of the long head of biceps. The anterior circumflex humeral also gives off a branch that 91  enters the intertubercular groove and pierces the head, (see figure I) This vessel, termed  31  the arcuate artery, supplies most of the blood to the head of the humerus and to the distal aspect of the supraspinatus tendon.  35,69,91  A branch of the thoracoacromial artery delivers  blood to the supraspinatus muscle then forms anastomoses with both circumflex humeral arteries. Rothman and Parkes' injection study also noted a previously undiscovered 91  artery, the suprahumeral artery. The suprahumeral artery "originates from the third 91  portion of the axillary artery and distributes blood to the anterior portion of the rotator cuff and lesser tuberosity of the humerus."  91  Figure 1. Graphic representation of the vascular supply of the humeral head. 1 = axillary artery; 2 = posterior circumflex humeral artery; 3 = anterior circumflex humeral artery; 4 = anteriolateral branch of the anterior circumflex humeral artery; 5 = greater tuberosity; 6 = lesser tuberosity; 7 = constant site of anteriolateral branch into bone 35  32  Several early studies have identified differences in the perfusion of blood vessels to muscles of the rotator cuff when compared to the tendons, especially the supraspinatus tendon. As early as 1934, Codman identified a "critical zone" in the supraspinatus tendon, which appeared anemic. Rothman and Parke identify the critical zone as an area usually in the distal part of the supraspinatus tendon and just proximal to its insertion as markedly undervascularized in relation to the remainder of the cuff. 91  Their study identified three patterns of tendon hypovascularity: supraspinatus tendon only (63%), both supraspinatus and infraspinatus tendons (37%), and subscapularis tendon (7%).  91  Considering that both the rotator cuff muscles and humeral head have both been found to be well vascularized, one must ask why does a hypovascular zone exist within the rotator cuff tendons? Several authors have attempted an explanation for this phenomenon. Mosley and Goldie relate the development of tendon degeneration to age related changes in bloodflow.Their study identified three sets of blood vessels in the cuff: osseous, muscular, and tendinous. Both the osseous and muscular vessels arise 69  directly from major arteries and are well vascularized. The tendons, however have "no main vessels concerned with its nutrition and must depend on anastomoses derived from muscular and osseous vessels." ' In the supraspinatus tendon, the main vascular supply 60 69  derives from vessels of the osteo-tendinous attachment (anterior circumflex humeral) and vesselsfromthe middle of the muscle (suprascapular). "With increasing age the number 58  of vessels towards the tendinous portion appear to decease somewhat." ' ' This may 41 58 69  explain why partial and complete tears of the rotator cuff occur almost entirely in older people.  33  The positioning of the upper extremity also plays a role in the hypovascularity of the supraspinatus tendon. Rathburn and MacNab found that when the shoulder is abducted, the supraspinatus tendon almostfillscompletely all of the vessels throughout the tendon to its point of insertion. ' However, when the arm is in a neutral position 41 84  (adducted and internally rotated), the supraspinatus tendon is stretched tightly over the humeral head. "The constant pressure exerted by the head of the humerus on the tendon of supraspinatus might wring out vessels in this area." The 'wringing out' of vessels 84  would prevent blood from reaching the distal portion of the tendon. The problem may be exacerbated by the configuration of the vessels within the tendon. Unlike round tendons, whose blood supply is largely derived from the paratenon and enters at intervals along the length of the tendon, in flat tendons, such as those comprising the rotator cuff, the vessels course through the whole length of the tendon, running mostly in a longitudinal direction. This disposition of the vessels renders them susceptible to traction and direct pressure. 84  Using a Scanning Electron Microscope, Ling et al. confirmed that" capillaries in the critical zone and muscle-tendon junction pursued a course parallel to the collagen fibres."  58  More recent studies have discovered other differences in the vascular bed of the supraspinatus tendon. Lohr and Uhthoff reported that "a distinct layering is present between the tissues of the bursal side and those of the articular side." Histologic 60  sections demonstrated that "the bursal side had an abundance of blood vessels, whereas the articular side was only sparsely supplied." This vascular pattern makes the articular 60  surface more susceptible to degenerative changes. Evidence of this theory is demonstrated by the development of "incomplete lesions of the tendon are found at the articular side in increasing numbers."  60  r  34  A similar pattern was described by Ling et al. They found that two levels of anastomoses between secondary arterioles were found: thicker anastomoses at thefirstlevel, with diameters of about 25 um and thinner anastomoses at the second level, with diameters of about 10 wm, virtually capillary in nature. Even more dramatic changes were noticed in the critical zone of the tendon. Scanning Electron Microscope revealed less anastomoses between secondary arterioles and a sparse capillary bed in the critical zone. Such a vascular pattern has a poor capability for formation of a collateral circulation when compressed and tends to cause ischemia. 58  Several studies employing dye injections in cadavers have demonstrated a hypovascular 'critical zone' in the supraspinatus tendon. Avascular regions have been associated with tendinitis, calcification and rupture of tendons. With the onset of tendon degeneration, secondary vascular phenomena were observed. Firstly, there was a reaction that appeared to be a foreign body inflammatory response with the development of vascular tufts of granulation tissue. 84  This reaction further impairs blood flow to the tendon, resulting in more degeneration of the tendon. "Relative avascularity of tendons over a prolonged period could be indicated as a sole cause of degenerative changes."  84  A new theory has recently developed that states tendon degeneration is not the result of a hypovascular tendon, but a hypervascular one. Studies using new techniques such as Laser Doppler Flowmetery (LDF) have found results that contradict those found in cadaver studies. Swiontkowski et al. report that "clinical observations show that the rotator cuff tendon tissue in the region of the impingement zone appears hyperemia"  102  Using an intraoperative LDF, they found that all patients who had rotator cuff tendinitis had significant bloodflowwithin the tendon. It seems reasonable to conclude that impingement produces a hyperemic response within the impingement zone of the tendon. 102  35  In a second study, Jensen et al. found "blood cell flux in the supraspinatus tendon at the microvascular level increased markedly during brief contraction." Large increases in 49  blood flow to the tendon seem unlikely if there is inadequate vasculature within the cuff. Damage to hyperemic tendon occurs "by resorption of damaged collagen fibres, which may lead to partial and complete tears." This study found no evidence of a 102  hypovascular zone. The occurrence of these zones in other studies is "probably an artifact of injection technique done in cadavers."  102  The vascular pattern of the rotator cuff remains a controversial issue in orthopedic research. Classic studies injecting dye into the arteries of cadavers have consistently found an avascular zone within the rotator cuff. These studies have been confirmed with Scanning Electron Microscopes of the cadaver dissections that have found poor vascular beds within the supraspinatus tendon. Recent studies and clinical observations by surgeons contradict these studies however. During surgery, the tendons of the cuff have been noted to bleed profusely when cut. Research employing Laser Doppler Flowmetery has found results that support the anecdotal evidence of the surgeons. These studies have demonstrated that the rotator cuff is well vascularized. At this time no definite conclusions on the rotator cuffs' role in the development of shoulder impingement can be made. Research seems to indicate that even if the cuff is hypo- or hypervascular, it may be at least partially responsible for the development of cuff lesions. Further research is required tofinda conclusive answer about the rotator cuff microvasculature.  36  ACROMIAL MORPHOLOGY Neer in 1972, noted a relationship between the shape and slope of the acromion and the development of rotator cuff lesions. Variations on the shape of the acromion were first classified by Bigilani et al. in 1986. Their study investigated the acromial shape in 140 elderly cadavers. They found three distinct types of acromion: type I is flat, II is curved, and III is hooked. They reported that 17.1% of the cadavers had a type I 13  acromion, 42.9% type II, and 39.3% type III. ' 57.7% of the cadavers has the same 12 13  acromial type in both shoulders. A second study by the same researchers evaluated the 12  supraspinatus outlet view of 200 patients with shoulder complaints. "The incidence of acromial types correlated closely with the anatomic study: 18% were type I, 41% were type II, and 41% were type III.  Type I  13  Type II  Type UI  Figure 2. Acromion types described by Bigliani and Morrison  Two further cadaver studies by Getz et al and Nicholson et al., both published in 1996, also examined the frequency of each acromial type. Getz et al. studied 394 cadaveric scapulas. The age of specimens ranged from twenty to eighty-nine years. They reported rates of 22.8%, 68.5%, and 8.6% for types I, II, and III respectively. They 36  found no relationship between acromial type and age.  36  37  Nicholson et al. published results comparable to those presented by Bigilani. They studied 420 cadaver scapulas. There were equal numbers of male and female and black and white subjects studied. They found an incidence of 32% for type I, 42% for type II, 76  and 26% for type III  76  Analysis of the data revealed no consistent statistically significant  impact of age on morphologic condition.  7 6  The results of Getz's and Nicholson's study  both indicate that acromial morphology does not change with age. The development of anterior spur formation however is age dependent. Nicholson et al. Reported that only 7% of subjects under age fifty had subacromial spurs while 30% of subjects over fifty demonstrated spur formation.  76  The presence of a downward sloping acromion of bone spurs can result in impingement and injury to the rotator cuff.  116  Bigilani et al. reported an association  between variations of the acromion and the development of rotator cuff tears. Their research found that 70% of the cadavers with type III acromions had rotator cuff tears while only 3% of type I were associated with tears. Variations in the shape and the 13  slope of the anterior acromion (type II and III) often results in a reduction of the subacromial space. The outcome is increased mechanical abrasion on the supraspinatus tendon and subsequent degeneration. Further investigation by Morrison and Bigilani discovered that this relationship is true only for patients with type III acromion. When "those patients with type I and II impingement are examined, there was no association between the type of acromion and diagnosis." In this group of patients, 19.4% had type 68  I acromion, 49.1% type II, and 31.4% had a type III.  68  In 1986 Hardy, Volger III, and White examined the radiographs of 36 patients with shoulder impingement syndrome. They found that 68% had subacromial bony spurs  38  and 66% had degenerative changes to the greater tuberosity.  Changes to the  acromioclavicular joint are also common. "Degenerative joint disease was seen in 66% of the radiographs. Other noted changes include "inferiorly oriented acromioclavicular osteophytes" in 32% of the subjects, degenerative changes of the lesser tuberosity in 29% of the cases, and 24% of patients had "small linear calcific densities in the region of the supraspinatus tendon." Narrowing of the joint space was noted in 21% of the 4 0  shoulders. 74% of the patients studied had between two and five of these abnormalities 40  noted on the radiographs. The results of this study suggest that subacromial spurring, inferior acromioclavicular joint spurring, and degenerative changes in the greater tuberosity develop in some individuals as part of the normal aging process. 40  These degenerative changes are felt to be major causes of the impingement syndrome. The degenerative spurring on the undersurface of the acromion may cause bony encroachment on the rotator cuff and play a major role in the pathogenesis of the impingement syndrome. It is not clear, however, whether episodes of impingement lead to the development of degenerative spurring or vice versa. 0  The coracoacromial ligament and acromioclavicular joint may also contribute to subacromial impingement. The influence of acromial shape has been well documented 68  in several papers. Variations in the coracoacromial ligament could have a similar influence on the supraspinatus tendon. "The fibers of the coracoacromial ligament fan out along the undersurface of the acromion and are extremely firm and unyielding."  94  Therefore, the ligament becomes a  39  prime suspect as an etiologic factor in chronic impingement syndrome, especially when bony or articular changes in the coracoacromial arch are not significant in roentgenograms. 94  A histologic examination of the ligament found "degenerative changes characterized by fibrillation of the collagen bundles, microtears, and fatty infiltration in the substance of the ligament." There was no indication however that these changes were extreme 9 4  enough to reduce the subacromial space. "It was concluded therefore, that the coracoacromial ligament was not a primary cause for impingement."  9 4  The shape of the acromion has frequently been described in the etiology of shoulder impingement syndrome. In addition to acromial shape, "congenital or traumatic anomalies of the acromion appear to be a separate risk factor for the etiology of the impingement syndrome and thus for eventual degeneration of the rotator cuff tendons."  101  The most common of these anomalies is the os acromiale. "The os acromiale is an unfused portion of the acromion which can lead to impingement syndrome in some athletes." An os acromiale is thought to be present in about 1.4% of the population. 101  101  Although the incidence of os acromiale is quite low, it may play a significant role in the development of shoulder impingement in athletes that have this anomaly.  40  RELATED RESEARCH Much of the literature published on shoulder impingement syndrome has been based on clinical observations of injured athletes by physicians or surgeons. At best, a few questionnaire studies were conducted to address the etiology of rotator cuff tendinitis. Recently however, some researches have examined the relationship between shoulder strength and flexibility and the occurrence of injury. To date, no definite cause and effect relationship has been made between test variables and the occurrence of injury. Greipp was one of thefirstresearchers to look for a cause and effect relationship between strength and flexibility and shoulder injury. He studied anterior shoulder flexibility and weight training patterns in 168 age group swimmers and used the data to predict future shoulder injury. "The data revealed a clear correlation between anterior shoulder inflexibility and subsequent shoulder pain." There was also a relationship 39  between heavy weight training and injury in male swimmers. It is theorized that increased shoulder pain "probably resultsfromtendon hypertrophy" Larger supraspinatus 39  tendons, having less clearance in the subacromial space, would be subject to greater mechanical impingement. Greipp therefore concludes "there is no evidence ... that weight training is beneficial as a preventive or therapeutic measure for swimmer's shoulder."  39  Warner et al. investigated the pattern of shoulder flexibility, laxity, and strength in athletes. This study quantitatively analyzed these characteristics infifty-threesubjects. Fifteen were asymptomatic volunteers, ten had impingement syndrome, and twenty-eight had glenohumeral instability. Patients with impingement demonstrated marked 110  limitations of active and passive internal rotation, and cross-chest adduction. The 110  41  authors attributed this phenomenon to a tight posterior capsule from reactive fibrosis tissue in the capsule as a result of repetitive microtrauma. ' Posterior capsule 15 110  contracture results in "superior humeral head translation during elevation of the shoulder in a forward plane."  110  Superior migration of the head further increases the amount of  mechanical impingement hence its common association with rotator cuff tendinitis. The impingement group also tended to have less translation in all planes when compared with both normal and instability groups. Rotator cuff muscle imbalances were also present 110  in the impingement group. This group had an "internal to external rotation strength ratio approaching two hundred percent or more"  1 1 0  This was significantly higher than either  the normal or instability group. The authors felt that this "represented a relative weakness of the external rotators despite failure to demonstrate consistent external rotation deficit."  110  However, they quote several authors that "have previously made the  clinical observation of external rotation weakness in this group of individuals."  1 1 0  A final  interesting note from this study was that repetitive overhead motions were responsible for almost all shoulder pain in these subjects. Control group Active internal rotation Passive internal rotation Passive external rotation  Impingement group  Instability  T6 46° 96°  T3 73° 108°  Cross chest adduction 20° Anterior drawer 0.3 0.4 Posterior drawer 0.5 0.2 Sulcus sign 0.6 0.4 Ratio IR/ERpeak torque 120% 165% @907sec Ratio IR/ER peak torque 150% 195% @1807sec Participation in overhead 31% 100% sports Table IV. Results from Warner, Micheli, Arslanian et al. 1990  group  52° 1.75 1.5 1.6 95% 120% 75%  42  From their literature review, Beach, Whitney, and Dickoff-Hoffman wrote that the two most commonly reported mechanisms of injury in the swimmers' shoulder were flexiblitiy and rotator cuff muscle imbalance. However, the literature is limited in 11  demonstrating a correlation between these factors and shoulder pain in competitive swimmers. Beach et al. studied shoulder flexibility, strength and endurance in thirty-two varsity level swimmers. Flexibility testing revealed "that the swimmers in this study were hypomobile in internal rotation compared with published standards." The data for 11  external rotation and abduction demonstrated hypermobility however.  11  The majority of swimmers in this study were hypermobile in shoulder abduction, external rotation and flexion compared with published norms, but not all swimmers experienced shoulder pain. Therefore, there was little correlation between hypermobility or hypomobility and shoulder pain in competitive swimmers. 11  "There was no significant correlation between external/internal rotation and abduction/adduction strength ratios and shoulder pain in competitive swimmers. Their 11  investigation did discover a relationship between the endurance ratios of the internal to external rotator cuff muscles and the incidence of shoulder pain. They found that as endurance ratios decreased the reported level of pain and dysfunction increased.  11  The results of the multiple regression analysis of external rotation and abduction endurance ratios showed a significant correlation to shoulder pain in competitive swimmers. This suggests that when evaluating swimmers, the clinician needs to be aware of the importance of assessing the endurance ratios of the shoulder abductors and the external rotators at 240°/sec, because as the endurance ratios decreased, the reported level of pain and dysfunction increased. 11  43  Shoulder ROM  Mean degrees  Correlation to shoulder pain  188° 62° 44°  Right 187° 59° 44°  Left 10° 16° 14°  Right 9° 14° 16°  Left .29 .21 .12  Right .17 -.01 -.15  141°  136°  10°  25°  .12  -.29  100°  101°  10°  11°  .08  .26  49°  45°  14°  12°  .01  -.03  196°  195°  14°  15°  -.03  -.27  Left Flexion Extension Horizontal abduction Horizontal adduction External rotation Internal rotation Abduction  Standard deviation degrees  Table V. ROM datafrom Beach, Whitney, andDickoff-Hoffman, Strength Ratio (a>. 60°/sec  ExternalAnternal rotation Abduction/adduction  Mean peak torque (ft-lbs.)  199211  Standard deviation (ft-lbs.)  Correlation to shoulder pain  Right  Left  Right (%) Left(%)  Right (%) Left  (/oj 71  70  10  9  -.03  -.12  57  56  11  12  .14  -.06  tan, 1992 11 Table IT. Strength ratio data fr om Beach, Whitney, and Dickoff-Hoffn Endurance Ratio @ 240°/sec  External rotation Internal rotation.  Mean peak torque (ft-lbs.)  Left (%) 80 106  Standard deviation (ft-lbs.)  Right (%) Left(%) 78 107  23 17  Correlation to shoulder pain  Right (%) Left 22 17  -.61 -.25  Right -.69 -.46  n  Table VII. Endurance ratio datafrom Beach, Whitney, and Dickoff-Hoffman, 1992  Differences in rotational strength and range of motion in subjects either with or without shoulder impingement syndrome was investigated by Leroux et al. The subject pool was composed of fifteen age matched, healthy volunteers, fifteen non-operated impingement subjects with type II impingement, and fifteen post-operative impingement subjects. They found differences in range of motion and internal to external strength 56  44  ratios between the two groups. The mean range o f motion in the normals was 127.3° while the impingement group averaged 1 1 3 ° .  56  The power tests found similar differences.  In the impingement group, mean values o f internal and external rotational strength for peak torque and average power demonstrated a significant deficit for the involved shoulder as compared with the control group with relative weakness in the internal rotators. 56  B o t h peak torque and average power ratios were significantly lower than the control group. These results differ from those o f Warner's et al. who found weak external rotators. Differences between the studies maybe the result o f different groups o f subjects. "The patients in Warner's study were younger and were sportsmen."  56  Group Tested Peak Torque (a 60°/sec Peak Torque (a 1807sec Control Dominant side 1.3 ± 0.2 1.4 ± 0.2 Non-dominant side 1.2 ± 0.3 1.3 ± 0.4 Impingement non-operated Involved side 1 ± 0.3 1.1+ 0.3 Uninvolved side 1+0.3 1+0.4 Impingement operated operated side 1.04 ± 0.4 1.2 ± 0.3 non-operated side 1.0 ± 0.2 1.1+ 0.2 Table  VIII. Internal/External  rotator ratio from Leroux,  Codine,  Thomas et al.  1994  The next study on swimmers shoulder was completed by Rupp, Berninger, and H o p f in 1995. They studied the strength and flexibility patterns in the shoulder o f twentytwo competitive swimmers and compared them to age and sex matched controls. The mean value for external rotation was 115° (right) and 113° (left) for the swimmers versus 80° (right) and 79° (left) for the controls. "The greater range o f motion is necessary to perform competitive swimming at a high level. Substantial laxity o f anteriorinferior capsuloligamentous structures is required to allow greater external rotation."  92  In  swimmers the ER/TR ratio for peak torque and total work was significantly lower than the  45  control group. At sixty degrees per second peak torque was 76.2% (right) and 68.3% (left) for the swimmers and 94.1% (right) and 82.6% (left) for the controls. Similar 92  results were identified at 180 degrees per second. The swimmers averaged 76.2% (right) and 80.1% (left) while the controls were 85.5% (right) and 96.6% (left). "There was no 92  significant difference in ER/IR ratio for peak torque and total work between those athletes who suffered from shoulder problems and those who did not."  92  Muscular imbalance does not seem to cause shoulder problems alone. Combined however, with attenuation of the capsuloligamentous structures, active stabilizers (muscles of the rotator cuff) may not be able to compensate for the insufficient passive stabilizers. 92  Another commonfindingin the swimmers was the presence of scapular winging and shoulder protraction. "These clinical phenomenon are indicative of dysfunction of 92  the scapular stabilizers." The authors concluded that competitive swimmers have: 9 2  • Greater range of motion especially during external rotation when the arm is abducted 90 degrees • Scapular winging and shoulder protraction in a high percentage of athletes • Positive impingement sign in a high percentage of athletes • Positive apprehension sign in a high percentage of athletes • Significantly lower external rotation / internal rotation (ER/IR) ratio than controls due to stronger internal rotators. 92  Swimmers Right  Swimmers Left  Control Right  External rot 69° 70° 59° (adducted) External rot. 113° 80° 115° (abducted) 92° Internal rot 96° 96° (adducted) Internal rot 74° 63° 74° (abducted) Table IX. Passive range of motion from Rupp, Berninger, & Hopf 1995  Control  Left  57° 79° 93° 60°  The occurrence of glenohumeral joint laxity in elite and recreation swimmers was investigated by Zemek and Magee in 1996. They tested thirty elite and thirty recreation  46  swimmers for shoulder laxity and general joint hypermobility. They completed an anterior, posterior, and inferior drawer and four general joint hypermobility tests on each subject. The elite swimmers were found to have greater laxity in both the anterior and inferior directions. There was no difference in posterior drawer. However, 119  119  there was considerable variation in posterior humeral head translation for different swimmers. Thus, the grading of the posterior drawer test proved to be insensitive to detect any clinically evident difference. 119  "The results of this study indicate that elite swimmers are more lax in their glenohumeral joints than are recreational swimmers." The pattern of increased laxity supports the 119  notion that increased laxity is acquired through intense swimming. However, the 119  results of general joint hypermobility assessment found that elite swimmers had more positive tests than the recreational. This suggests an inherent basis for the excess 119  laxity.  119  "It is proposed that both factors contribute the greater laxity demonstrated by  elite swimmers, although the relative importance of each factor is still unclear."  1 1 9  A greater number of elite swimmers demonstrated a positive history of shoulder overuse dysfunction.  119  "The finding of greater glenohumeral joint laxity in the elite  group provide indirect support for the hyperlaxity theory for shoulder overuse dysfunction in these athletes."  119  Bak and Fauno report the main factors associated with shoulder pain in swimmers 7 21 53 64 87  to be swimming experience, faulty stroke technique, and training errors. ' ' ' ' "The cause of pain is a combination of overuse (repetitive microinjuries) and overload (e.g. sudden increases in training load)." Biomechanical research studies suggest that significant changes in muscle activity and strength, particularly of the shoulder rotators and scapular stabilizers, play an important role in the development of shoulder problems in the overhead athlete. The muscular imbalance is confined not only to the 7  47  glenohumeral joint but to the scapular stabilizers as well. An imbalance or weakening of the scapular muscles makes anterior translation of the humeral head possible, provoking impingement. 7  Bak and Faounos' investigation of thirty-six clinically injured swimmers produced the following results. The painful arc of elevation was seen in only 27% of the painful shoulders. Bak and Fauno speculate the painful arc was "rarely seen because a 7  painful arc is incompatible with swimming." A positive Hawkins Impingement test was 7  discovered in thirty-nine of forty-nine painful shoulders andfiveof twenty-three asymptomatic shoulders. A Neer Impingement test was positive in nineteen of forty7  nine injured shoulders and no non-injured. An anterior drawer of +2 or more was 7  discovered in nineteen painful shoulders (39%) and a sulcus sign of +2 in sixteen (33%).  7  Finally, lack of coordination at the scapulohumeral joint was seen in 33% of symptomatic shoulders compared with 9% of non-symptomatic.  7  There is increasing evidence that scapulothoracic instability plays a role in the overhead athletes' shoulder problems. Dynamic electromyographic recordings reveal significantly lower serratus anterior muscle activity in swimmers with shoulder pain compared with swimmers with no pain. The trapezoid and serratus anterior muscles seem to play the major roles in axioscapular muscle coordination and evidence is being provided that shoulder instability is associated with fatigue and dysfunction of these muscles. 7  Potential causes of shoulder injury were investigated by Bak and Magnusson in 1997. They examined the relationship between shoulder strength and range of motion in seven injured swimmers with type I or II impingement and eight swimmers with no history of shoulder pain or instability. Comparison of an injured shoulder to the contralateral shoulder found "no difference in external rotation between painful and painfree shoulders." There was a "trend towards lower concentric (p = 0.06) and eccentric (p 8  = 0.07) internal rotational strength in painful shoulders." Bak and Magnusson also 8  48  found no difference in conventional concentric and eccentric external rotation to internal rotation (ER:IR) strength ratios.  8  Comparing injured swimmers to non-injured, Bak and Magnusson found no difference in external torque or internal rotational strength although the difference approached statistical significance for concentric internal rotation (p = 0.053) and eccentric internal rotation (p = 0.13) strength.  8  Symptomatic shoulders were found to have significantly greater concentric (p = 0.02) and eccentric (p = 0.02) E R : I R strength ratios than pain-free shoulders.  8  Symptomatic shoulders in this study displayed decreased internal rotational strength compared with non-symptomatic shoulders. "Dynamic E M G analysis o f shoulder girdle 8  muscles in swimmers shows no significant differences in the strong extraarticular internal rotators and adductors o f the humerus." Significant differences were noted in the rotator 8  cuff muscles and the scapular stabilizers as a result o f shoulder p a i n . Bak and 8  Magnusson conclude that this study "does not give a final proof o f the cause and effect relationship o f shoulder strength to pain."  8  Internal range o f motion was also reduced (although not significantly) in painful shoulders compared with pain-free (p = 0.14). N o differences in external range o f 8  motion were detected. The increased range of motion in all swimmers "might be explained by a physiologic adaptation to repetitive stress o f the anterior labrum and capsule."  8  External rotation Internal rotation Total ROM Concentric ER:IR ratio Eccentric ER:IR ratio Functional ER:IR ratio  Injured side  Healthy side  Control group  110°  110°  106°  66°  68°  73°  176°  177°  179°  0.83  0.78  0.66  0.71  0.78  0.67  1.07  0.89  0.86  Table X. Rotation and ER.IR ratio results from Bak and Magnusson, 1997  49  II. HYPOTHESIS The literature on shoulder impingement syndrome reports several etiologic factors that may lead to the development of shoulder injuries in swimmers. Common factors include: hypovascularity, acromial variations, rotator cuff muscle imbalances, overwork, joint laxity, scapular dysfunction, and posterior capsule inflexibility. (See figure III) The role of hypovascularity is unclear in the development of shoulder impingement syndrome. Many classic studies have cited the presence of an avascular critical zone in the supraspinatus tendon. Several studies using arthrograms in cadavers and live subjects support this notion. However, some current research using Laser Doppler Flowmetery suggests that the supraspinatus tendon is actually hypervascular. This study combined with reports that the supraspinatus tendon bleeds profusely during operations imply that avascularity may not be a key factor in the development of impingement syndrome. The influence of acromial variations in the development of shoulder impingement syndrome is also unclear. Many studies on acromial variations and osteophyte formation have been performed on cadavers. Controversy exists in the literature over whether abnormality of the acromion results in rotator cuff tendinitis or if chronic abrasion of the cuff on the undersurface of the acromion causes changes to its anatomy. In a young athletic population, acromial morphology probably plays little or no role in the development of shoulder impingement syndrome. Competitive swimmers have a tendency to develop rotator cuff muscle imbalances. Some swimming strokes, namely freestyle and butterfly, selectively train the internal rotators. If imbalances exist between internal and external rotators, the  50  glenohumeral joint will not be adequately stabilized.  The resulting instability will cause  excess translation of the humeral head increasing the likelihood of impingement occurring. Overwork is another significant factor in the development of impingement syndrome in swimmers. As mentioned previously, a competitive swimmer may complete up to 750 000 shoulder rotations in an average season. With each rotation, the humeral 63  head may mechanically impinge the supraspinatus tendon on the undersurface of the coracoacromial arch. The excessive amount of work performed by swimmers can lead to fatigue of rotator cuff muscles and laxity of the joint capsule and ligaments. Once the stabilizers begin to fail, the humeral head slides into the coracoacromial arch, mechanically impinging the supraspinatus tendon and subacromial bursa. Excess laxity of the glenohumeral joint also contributes to the development of shoulder impingement syndrome. Swimmers tend to place a lot of emphasis on shoulder flexibility exercises. Over-stretching of the shoulders coupled with overwork results in laxity of the shoulder. As mentioned previously, joint laxity allows excess translation of the humeral head. A final factor related to the development of rotator cuff tendinitis is posterior capsule inflexibility. Tightness of the posterior capsule causes anterior translation of the humeral head into the undersurface of the acromion. Either inadequate stretching or a reflexive contracture may result in an inflexible posterior capsule.  51  Glenohumeral Joint Laxity Hypovascularity  Acromial Morphology  SHOULDER IMPINGEMENT SYNDROME Overwork  ^  Training Errors Rotator Cuff Muscle Imbalances  Posterior Capsule Inflexibility  Scapulothoracic Dysfunction  Secondary factors Primary Factors Figure 3. Hypothetical combination ofEtiologic Factors in the Development of Shoulder Impingement Syndrome In the young competitive swimmer only certain factors play a role in the development of impingement syndrome. Congenital variations in the acromion may exist in younger swimmers however, this phenomenon has not been often cited as a major contributing factor. Therefore, in young competitive swimmers, overwork, joint laxity, posterior capsule inflexibility, rotator cuff muscle imbalances, and possibly scapular dysfunction are most likely the critical elements in the development of shoulder impingement syndrome. The focus of this investigation is to determine an answer to the following question; Is there significant difference in workload, joint laxity, scapular dysfunction, acromial morphology, posterior capsule tightness and / or rotator cuff muscle imbalances between swimmers with or without shoulder impingement syndrome?  52  The following results are expectedfromthis study: 1. There will not be a significant difference in overwork between groups. All swimmers in general overwork in their preparation for competition. Therefore, athletes competing at similar levels will have similar training patterns. An association may exist between recent increases in workload and the development of impingement. If an athlete is not properly conditioned, rapid increases in workload may accelerate tissue degeneration leading to the formation of lesions. 2. Due to the nature of swimming strokes and the emphasis placed on shoulder flexibility, all swimmers will display some degree of joint laxity. Excess laxity permits upward movement of the humeral head, which results in mechanical impingement of the head on the undersurface of the coracoacromial arch. Therefore, athletes with excessive laxity will have a greater incidence of shoulder impingement syndrome. Hypermobile shoulders joints in these athletes maybe the result of either too much emphasis placed on stretching or an individual generalized ligament laxity. 3. Acromial morphology will play no role in the development of shoulder impingement syndrome in most young, athletic subjects. The formation of osteophytes usually takes many years and the results of the bone spurs will not usually noticed until later in life. 4. Scapular dysfunction resulting from weak serratus anterior muscles may be evident in some subjects with shoulder impingement syndrome. 5. Posterior capsule inflexibility will differ between groups. Many swimmers with impingement syndrome will present with inflexibility in the posterior aspect of the joint capsule. Tightness in the posterior capsule forces the humeral head into the acromion, increasing the amount of mechanical impingement.  53  6. Swimmers diagnosed with shoulder impingement syndrome will have strength imbalances in the rotator cuff muscles. Competitive swimmers typically over-develop the internal rotators of the shoulder. Both the freestyle and butterfly strokes require a swimmer to perform internal rotation. If corrective measures are not taken, a dryland strengthening program for example, the internal rotators will become progressively stronger while the external will remain at their pre-training level. The imbalance between the strong internal and weak external rotators adversely affects the stability of the shoulder. The injured athletes then will have developed large muscle imbalances resulting in shoulder instability. The injured athletes will also have a reduction in the endurance ratio of the external rotators. 7. There are a few common swimming stroke technique errors that may exacerbate the development of shoulder injury. Athletes that breathe more frequently to one side, those with less body roll, and those with lower elbow recovery height while swimming frontcrawl are more likely to become injured.  54  III. METHODS Athletes from several different clubs acted as subjects for this study. Competitive age group swimmers from the University of British Columbia Swim team, Simon Fraser University Swim and Dive team, and local swim clubs were recruited to participate in this study. Age group swimmers were categorized based on the medical history of their shoulders, as 1 acutely injured, 2 history of injury, 3 non-injured. Masters athletes with chronic shoulder injury were also recruited and form the fourth test group. Age group swimmers between the age of thirteen and twenty-five were accepted into the study. Masters athletes were thirty years or older. To be classified as injured, an athlete had to have current shoulder pain that interfered with training or competition, or shoulder pain lasting several hours following practice and a positive diagnosis of shoulder impingement syndrome. Injured subjects were examined by Dr. Jack Taunton of the Allan McGavin Sports Medicine Centre to confirm the presence of shoulder impingement syndrome. Only athletes with Stage I and II shoulder impingement syndrome that have exhibited symptoms for a minimum of two months were accepted into the injured group. Diagnosis of rotator cuff injuries will be made using two clinical examinations. The impingement sign consists of forcibly flexing the shoulder forward, pushing the greater tuberosity against the anterior-inferior surface of the acromion. A positive impingement sign is indicated when this position produces 15  pain. A positive Hawkins impingement test is obtained when forced internal rotation with the shoulder flexed to 90° produces pain.  15  Athletes in the history group had to have shoulder pain that interfered with training or competition or shoulder pain lasting several hours following practice as a  55  result of competitive swimming. Many of these athletes had been diagnosed with rotator cuff injury by other doctors or physiotherapists, however the diagnosis was not confirmed in all subjects. Based on their age and involvement in swimming, shoulder impingement syndrome is the most likely cause of the previous dysfunction. However, the only conclusion that can be drawn regarding this group is that they have a history of shoulder pain since no attempt to diagnose previous injury was made. The non-injured group was composed of athletes that claim to never of had serious or interfering shoulder pain resulting from competitive swimming. Several different tests will be performed to determine if there are differences in overwork, stroke mechanics, scapular function, glenohumeral laxity, acromial morphology, posterior capsule tightness, and rotator cuff muscle imbalances. The tests will include surveys, and strength and flexibility measurements. 1.  The amount of work performed by each athlete will be determined from a  questionnaire. The survey will also collect personal information from the athletes, (see table II) Questions in the survey will focus on frequency, intensity, and duration of exercise, both in and outside of the pool. Training patterns, such as the amount of time spent on each stroke, use of hand paddles, and stretching routines will also be investigated. Finally, personal information will include information such as age, sex, and previous medical history. A second questionnaire will be given to the coaches to assess each athletes stroke mechanics to assess if variations in stroke play a role in the development of impingement.  56  Personal Information • • • •  age sex previous medical history swimming related injuries previous shoulder injuries history of shoulder pain years competitive swimming  Workload hours/day; days/week, months/year meters/session & week hours dryland training/week hours warm-up rating of perceived exertion (RPE)  Train ing information stroke specialization race distances stretching routines hand paddle use amount of time spent training in each stroke  Table XI. Survey Information  2.  "Shoulder instability can be anterior, posterior or inferior (multidirectional),  subtle or obvious." Laxity of the glenohumeral joint will gauged using several 1 5  diagnostic tests. •  Active internal rotation will be assessed as the highest vertebral level the subjects thumb can reach when the arm is moved behind the back.  •  Passive internal and external rotation, will be measured with a goniometer with the subjects in a supine position. Internal rotation will be measured with the shoulder abducted to 90° and the elbow flexed to 90°. External rotation will be measured with the shoulder abducted to 90° and 0°. Range of internal and external motion will be measured with a goniometer. Cross-chest adduction will be conducted to assess posterior capsule flexibility. This assessment will be conducted with the subjects supine and the shoulder flexed to 90° and 0° of elbow flexion. From this position, the subject will move his/her arm into cross-body adduction. Care is taken to ensure that the scapula and shoulder remain in contact with the bench. The amount of flexibility in the posterior capsule will be measured as the distance between the cubital fossa and the anterior aspect of the opposite shoulder.  57  •  Anterior and posterior sulcus (drawer) tests will be conducted to assess joint laxity. This involves "applying force to the elbow to reduce the humeral head, the anterior and posterior force is applied to the humerus and translation graded." The test will 100  be scored in the following manner: +1 implies that the humeral head translated further than the contralateral shoulder; +2 means that the examiner is able to sublux the humeral head over the glenoid rim but it spontaneously returns; +3 indicates that the examiner can lock the humeral head over the glenoid rim.  100  One limitation to using these methods to assess joint laxity is the subjective component of the results. Using a three point grading system does not provide a great deal of sensitivity and there is no 'ruler' that can be used to determine exact measurements. However, these are the standard exams used in the medical profession and have been accepted within the community as good measurement tools. Additionally, since only one examiner will test all subjects, reliability should be high. •  Inferior instability will be measured with the sulcus sign. In this test the shoulder is held in "zero degrees abduction, neutral rotation, and neutral flexion/extension."  100  Inferior traction is then applied to the shoulder by grasping the elbow. "Excessive inferior translation is manifested by a widening of the subacromial space between the acromion and humeral head. Grading of the sulcus sign will be measured in 15  millimeters. A ruler will be used to measure the distance between the acromion and humeral head. •  Finally, the athletes will be assessed to determine if they have positive impingement tests. The impingement tests involvefirststabilizing the scapula. The Neer Impingement test consists of forcibly flexing the subjects shoulder through a full  58  range of motion. The Hawkins Impingement test involves elevating the shoulder to 90° then forcibly internally rotating the arm. If either action results in a recurrence of symptoms it is considered a positive impingement test. The long head of the biceps will be tested in a similar manner. Subjects will flex their shoulder to 90° and 0° of elbow flexion. They will then be instructed to hold this position while the examiner applies downward pressure to the arm. A positive Speed's test will be noted if this action results in anterior shoulder pain. •  Each subject will also complete a test of generalized joint laxity. Generalized joint laxity will be measured as the distance between the distal end of the thumb and the radius during maximal wrist flexion. A ruler will be used to measure distance in millimeters. This test will help determine if congenital laxity or excessive stretching is responsible for the shoulder instability apparent in some swimmers.  3.  Function of the serratus anterior muscle will be assessed by having the subjects  perform a wall push-up. If this muscle is functioning properly, the scapula will remain in contact with the thoracic wall. Scapular winging will identify those with poorly functioning serratus muscles and therefore scapulothoracic dysfunction. This test will be scored as either positive or negative. A positive score is given if gross winging of the scapula occurs during the push-up. Winging is considered gross if the scapula pulls well away from the thorax. The serratus anterior will be tested both before and after strength testing of the rotator cuff muscles. 4.  Acromial morphology will be determined from radiographs of the injured  athletes' shoulders. The Supraspinatus Outlet view described by Aoki et al. will be used to assess the slope of the acromion. This is a "lateral scapular roentgenogram view in the  59  scapular plane at one meter through a fluoroscope so that the clear shape of the supraspinatus fossa and spine of the scapula were reflected as a single line of arc." This 5  view allows clinical measurement of the angle of the acromion. Bigliani's method of categorizing the acromions was used. A relatively flat acromion is classified as a type I. An acromion with a curvature in the posterior third is classified as type II. Finally, if a curvature in the anterior third is present the acromion is classified as type III. All of the radiographs were examined by one radiologist. A true A-P and Axillary view x-rays will be taken to determine if other pathologies are present in the shoulder. The A-P view can be used to examine the acromioclavicular and glenohumeral joints and surrounding tendons for abnormalities.  54  The Axillary view is "obtained primarily to evaluate for an Os Acromiale."  54  X-rays will not be taken of the normal subjects. The injured subjects will be compared to average published values. 5.  The final measurement will entail determining if there is a difference in internal  and external shoulder rotation strength. This will be accomplished using a Cybex II Isokinetic Dynamometer and upper body exercise and testing table (UBXT) (Lumex Corp. Ronkonkoma, NY.) in the Buchannan lab at the University of British Columbia. Prior to testing on the UBXT subjects will warm-up for five minutes on a cycle ergometer and perform fifteen to twenty repetitions of internal/external rotation using a stretch cord. After warm-up, subjects may perform shoulder-stretching exercises until they are ready to begin testing. Subjects are positioned on the Cybex in a seated position. Strength of the rotators will be measured in one position. The shoulder abducted to between 20° and 30°and the  60  elbow is flexed to 90°. Peak torque and total work will be measured at velocities of 90° and 180° per second. Subjects begin with five submaximal contractions at a velocity of 90° per second to warm-up and familiarize themselves with the test. Subjects then complete three maximal contractions each at 90° and 180° per second to determine peak torque. Muscular endurance will be assessed using the test protocol described by Beach, Whitney, and Dickoff-Hoffman in 1992. Each athlete will performfiftyinternal/external rotations at 240° per second. The strength decrement (endurance ratio) will be 11  calculated by dividing the mean torque of the highest three consecutive contractions performed during thefirstten repetitions divided by the lowest three consecutive contractions performed during the last ten repetitions. The right arm is always tested first. After a short rest period, the same test protocol is used for the contralateral limb. The results of the Cybex testing will be used to determine maximal internal and external rotation (torque), total work, internal to external strength ratio and torque per body weight ratio.  61  STATISTICAL ANALYSIS Several different statistical procedures were utilized to evaluate the data from this study. First, personal data and training patterns are presented as means for each group. Weight training, dryland activities and stretching patterns are presented in graphs for each group. Differences between the three age-group swimmer categories were evaluated using multiple analysis of variance (MANOVA) to assess groups of variables and subsequent analysis of variance (ANOVA) for each individual factor. The Hotelling's Trace value was used as the MANOVA score. The confidence interval was set at ninety five percent (p < 0.05) for results to be considered significant. MANOVAs were performed on the following variable groups: 1. Training information including duration, type of training and frequency of each stroke 2. Competition data including competition level, race specialization and race distance 3. Pain measurements composed of history of shoulder pain, current shoulder pain, and level of current shoulder pain 4. Impingement test data including Neer and Hawkins Impingement tests and Biceps Resistance test 5. Flexibility measurements including cross chest adduction and internal/external rotation range of motion data 6. Joint laxity measurements including anterior and posterior drawer, sulcus, and generalized joint laxity data 7. All strength measurements for both the left and right hand sides 8. Stretching data including age stretching began, frequency and duration of stretching A Student-Newman-Keuls post hoc test was completed after the ANOVA testing. The Newman-Keuls post hoc was used to evaluate the differences between the three injury categories. No post-hoc testing was conducted for the comparison of the masters and non-injured group as there are only two groups.  62  The data on the masters athletes was analyzed using the same MANOVA and ANOVA format. No non-injured masters athletes were tested to serve as controls in this study. Instead, the injured masters athletes were compared to the non-injured age group swimmers. The results of this analysis was compared to the analysis of age group < swimmers to determine if a similar trend in etiologic factors exists in both the age group and masters injured athletes. Finally, logistic regression analysis will be conducted on the variables that are statistically significant between groups. Multiple regression analysis requires continuous data for an adequate analysis. However, subjects in this study are classified as either injured or non-injured. Logistic regression was developed to analyze data with discrete values and is used extensively in biomedical research. Factors that are found to be significant from the ANOVA testing will entered into the logistic regression analysis. The resulting data will be used to develop a mathematical model. This model can be used to identify individuals that are at risk of developing shoulder impingement syndrome so that appropriate action can be taken to prevent injury from occurring. Subjects were classified as either injured (masters and age-group) or non-injured (non-injured and history groups). The logistic regression was run to determine differences between the two groups. The enter method was used to evaluate the data. Significance was set ap < 0.05.  63  IV. RESULTS One hundred seven subjects volunteered to participate in this study. The number of subjects in each group is: • • • •  twenty acutely injured, age group swimmers with type I or II shoulder impingement syndrome (injured group) thirty-five age group swimmers with a history of shoulder injury (history group) forty age group swimmers with no history of shoulder injury (non-injured group) twelve masters athletes with chronic type I or II shoulder impingement (masters group)  The average age for each group was; 16.50 years for the injured, 15.98 for the noninjured, 16.71 for the history, and 44.08 for the masters. There are eleven males and nine females in the injured group, twenty-two males and eighteen females in the non-injured group, fifteen males and twenty females in the history group and six males and females in the masters group. Eighty-eight percent of the subjects were right handed. Of the injured swimmers, two had right shoulder injury, six had left side, and twelve had bilateral rotator cuff injury. The masters group was composed of six right side injuries, four left side injuries, and two with bilateral shoulder injury.  Group  Age (years)  Males (%)  Female (%)  Right Handed (%)  Injured Non-injured  16.50 + 2.8 15.98 + 2.3 16.71+2.2 44.08 + 7.05  11 (55%) 22 (55%) 15 (42.86%) 6 (50%)  9 (45%)  16 (80%) 35 (87.5%) 31 (88.57%) 12 (100%)  History Masters  18 (45%) 20 (57.14%) 6 (50%)  Table XII Subject Personal Data Thefirstseries of MANOVA tests were performed on the age group swimmers data. The initial analysis investigated differences in training patterns among the age group athletes. The training volume and duration for each athlete was studied with a questionnaire. Questions included hours of dryland training, amount of training per year, week and day, meters of swimming per week and practice, meters of frontcrawl per 64  week, and the percentage amount of time each stroke is practiced. The Hotelling's Trace MANOVA was non-significant (p = 0.631) for the training data. The ANOVA analysis also failed to find a significant difference for any variable.  Dryland training Months/year Days/week Hours/day Meters/practice Meters/week m. frontcrawlAveek % frontcrawl % backstroke % breaststroke % butterfly  Injured  Non-injured  History  P value  3.51±1.45 10.65+1.04 5.78+90 3.01+96 4722.6+1766.6 40571+18904 25917+11803 58.2%±18.9% 17.1%±12.7% 11.4%±7.4% 12.1%±11.2%  3.83+1.38 10.55+1.14 6.06+46 3.11+94 5442.3+1059.8 46765+12377 28671+12819 56.9%±18.2% 19.3%±15.2% 9.2%±7.7% 14.0%±9.0%  4.04+1.36 10.55+1.04 5.97+17 2.88±.85 5172.6+1045.2 44392+13714 29354+12592 63.3%±16.7% 16.6%±10.3% 10.3%±7.7% 9.7%±6.5%  0 730 0 912 0 670 0 830 0 324 0 468 0 457 0 186 0 556 0 970 0 407  Table XIII. MeaniS.D. training volume for age group swimmers Subjects also answered questions on the distance and stroke they most commonly raced, current best times for one hundred and two hundred meter freestyle, and years of competitive swimming. There was no significant difference in years of competitive swimming, or one hundred and two hundred meter freestyle times within the age-group swimmer categories (p = 0.829) No difference was found in an analysis of competition level (p = 0.979), stroke specialization (p = 0.902), or race distance (p = 0.365). The MANOVA for the entire data set was also non-significant (p = 0.932). The competition level and race stroke specialization are broken down and presented in tables XIV and XV. The stoke specialization percentages do not add up to one hundred as several athletes race in more than one stroke. Years of swimming, frontcrawl times, and race distances are presented in table XVI.  65  Local Provincial College/University National International  Injured 0 (0%) 7 (35%) 1 (6%) 9 (45%) 3 (15%)  Non-injured 0 (0%) 13 (32.5%) 3 (7.5%) 20 (50%) 4 (10%)  History 0 (0%) 11 (31.43%) 4(11.43%) 16 (42.86%) 3 (8.57%)  Table XIV. Competition level of age group swimmers Frontcrawl Backcrawl Breaststroke Butterfly Individual medley  Injured 10(50%) 4(20%) 0(0%) 7(35%) 3(15%)  Non-injured 20(50%) 9(22.5%) 8(20%) 7(17.5%) 5(12.5%)  History 22(62.86%) 9(25.71%) 5(14.29%) 10(28.57%) 6(11.43%)  Table XV. Stroke specialization for age group swimmers.  Years competitive swimming 100 meter freestyle time (sec) 200 meter freestyle time (sec) Race Distance (meters)  Injured 8.65 ± 4 . 4  Non-injured 7.61 ± 2.3  History 7.81+2.65  /; value 0.363  59.52 + 8.5  60.30 + 5.1  60.89 + 5.9  0.871  131.00+13.2  132.26 ± 17.3  130.43 ±12.3  0.858  222.88 + 231.4  376.20 + 425.9  329.79 + 461.8  0.365  Table XVI. Mean±S.D. age group swimming history and current best times As one would probably assume, the injured athletes were significantly higher in the history of shoulder pain, the incidence of current shoulder pain, and rating of current shoulder pain (MANOVAp < 0.001). The interesting finding is the high number of noninjured swimmers with a history of shoulder pain. The Newman-Keuls test indicates that there is no difference between the non-injured and history groups for presence of current pain (p = 0.630) or current pain rating (p - 0.676). History of shoulder pain and current shoulder pain is presented as the number and percentage of individuals reporting pain at some point in time. The pain rating was determined from a seven point rating scale.  66  Pain history Current pain Pain rating  Injured 20(100%) 18(90%) 3.55+1.61  Non-injured 29(72.5%) 6(15%) 0.68+81  History 35 (100%) 7(20%) 0.79+64  p value < 0.001 < 0.001 < 0.001  Table XVII. Current and past age-group shoulder pain  The injured athletes had significantly more positive Neer and Hawkins Impingement and Speed's tests (p< 0.001). The number and percentage of subjects in each group reporting positive impingement tests are listed in table XVTII. The injured group consisted of eighteen left and fourteen right side injuries (two right, six left, twelve bilateral). The Neer Impingement test diagnosed sixteen (89%) of the left side injuries and fourteen (100%) of the right. The Hawkins Impingement test was positive in nineteen (105%) left injured shoulders and twenty (142%) right injured shoulders in the injured group. Several false positives were recorded with each test. Twenty-three (57.5%) of the Hawkins tests were recorded as positive in the non-injured group. The Neer Impingement test produced fewer false positives in the non-injured group (left 15%, right 25%). The occurrence of positive Speed's tests was non-existent for the non-injured group. Six (33%) left and four (28%) right injured shoulders were found to have positive biceps resistance tests. This is an indication that the majority of injury in the age group swimmers' shoulder is isolated to the supraspinatus and not the long head of the biceps. The Newman-Keuls post-hoc testing found no significant difference between the noninjured and history group for left and right Neer Impingement tests, left Hawkins impingement test, and Biceps Resistance tests. The history group had a high incidence of right side Hawkins Impingement tests. Consequently, the history group had significantly more right side Hawkins Impingement tests than the non-injured. There is a trend towards  67  the injured group having more right side Hawkins Impingement tests than the history group (p = 0.066). Injured  Non-injured  History  16 (80%) 6(15%) 9 (25.7%) Left Impingement Test 10 (25%) 9 (25.7%) Right Impingement Test 14 (70%) Left Hawkins Test 19 (95%) 23 (57.5%) 23 (65.7%) 20 (100%) 23 (57.5%) 28 (80%) Right Hawkins Test Left Speed's Test 6 (30%) 0 (0%) 3 (8.6%) 4 (20%) Right Speed's Test 0 (0%) 3 (8.6%) Table XVIII. Results of age group impingement and biceps resistance tests.  p value < 0.001 0.001 0.010 0.001 0.001 0.071  A significant difference in shoulderflexibilitywas discovered between the age group swimmers. A significant Hotelling's Trace MANOVA (p = 0.003) was calculated for the total assessment of shoulder flexibility. There does not appear to be any difference in external rotation between the groups although there is a trend towards significance (p = 0.087) for right external rotation at zero degrees abduction. There appears to be a trend towards decreased passive (right p = 0.101) and active (left p = 0.124, rightp = 0.019) internal rotation in the injured athletes. There is a distinct decrease in the flexibility of the posterior joint capsule, measured by cross chest adduction, of the injured swimmers (leftp < 0.001, rightp = 0.042). The results of the post-hoc tests for cross-chest adduction indicate that there is no difference between the non-injured and history group (p = 0.543) on the left side. The history group is not significantly different from the injured (p = 0.296) or non-injured group (p = 0.134) for the right side. The increased incidence of posterior capsule inflexibility in the history group's right shoulders may partially account for the high number of Hawkins Impingement tests in this group. The Newman-Keuls results for the passive internal rotation also indicate a trend towards reduced range of motion in the injured group for the left side (p = 0.088). No  68  significant difference was noted on the right side between the history and non-injured group (p = 0.293). Active abduction was essentially equal in all three groups with only two injured athletes displaying a slight decrease in range of motion (mean injured = 179.625°, noninjured = 180°, history = 180°). The classic painful arc of abduction was not seen in these athletes. Pain was not noticed between the range of 60° to 120° as has been reported in the literature but was seen to occur in a range from 160° to 180° (mean level that pain was initially felt: injured group left = 173.125°, right = 171°; mean for the history group left = 175°, right = 172.25°). In the injured group, painful abduction was present on the left side in eight shoulders and five right shoulders. The history group had two left shoulders and fourrightshoulders with pain during abduction. Painful abduction was not noted in the non-injured group. Injured  Non-injured  Left passive int rotation 100.2°±22.0 107.9°±19.9 107.9°±20.7 Right passive int rotation 97.6°±23.3 Left passive ext rotation @90° 111.0°±17.4 115.8°±12.2 Right passive ext rot @90° 115.1°±12.3 118.7°±14.3 96.0°±23.5 100.5°±16.2 Left passive ext rotation @0° Right passive ext rot @0° 90.9°±23.4 99.5°±15.7 14.9±1.5 15.8±1.3 Left active int rotation Right active int rotation 14.3+1.8 15.4+1.9 Left cross-chest adduction6.4 cm±5.3 2.6 cm±2.8 Right cross-chest adduction 6.5 cm±4.6 3.7 cm±3.3 Table XIX. Means±S.D.  History  p value  110.2°±20.4 107.8°±20.7 115.9°±15.4 120.9°+13.5 96.5°±18.2 94.0°+20.2 15.8+1.9 15.2+1.6 3.0 cm+3.3 5.3 cm±3.9  0.217 0.101 0.526 0.268 0.426 0.087 0.124 0.019 < 0.001 0.042  for age group flexibility data  The examination of the glenohumeral ligaments found that injured swimmers have more joint laxity than their non-injured counterparts (p = 0.004). Injured athletes have significantly greater anterior drawer (left p = 0.003,rightp = 0.012) and inferior  69  sulcus (left p = 0.007,rightp < 0.001). There was no significant difference in posterior drawer. Post-hoc testing reveals that the anterior drawer results for the history group are not significantly different from the non-injured group on the left side (p = 0.462) or the injured on the right side (p = 0.894). It is expected that the history group would have similar laxity patterns as the injured group. The non-significant finding for the left side anterior drawer is probably a result of limitations in the grading system. Results for the sulcus tests indicate that all three groups are significantly different on the right side and a trend towards significance on the left (history compared to injured p = 0.069). The athletes were also assessed for generalized joint laxity by measuring the amount of flexion available in the right wrist. Generalized joint laxity is considered positive if the distance between the thumb and wrist is less than four millimeters. Between 37% and 45% of the athletes could be considered to have generalized joint laxity. The second generalized laxity measurement is the average distance between the thumb and wrist expressed in millimeters. There was no significant difference on either measure between the three groups.  Left anterior drawer Right anterior drawer Left posterior drawer Right posterior drawer Left inferior sulcus Right inferior sulcus Generalized Joint Laxity Generalized Joint Laxity 2  Injured  Non-injured  History  /; value  0.90+72 0.75+79 1.05+83 0.85+88 1.28+68 1.33+65 8 (40%) 22.50+23.1  0.33+47 0.35+48 0.68+69 0.65+70 0.83+40 0.75+45 18 (45%) 20.88+24.9  0.44+67 0.73+63 0.69+75 0.86+72 1.03+51 1.03+47 15 (37%) 21.14+19.7  0.003 0.012 0.147 0.424 0.007 < 0.001 0.789 0.965  Table XX. MeansdS.D. for age group joint laxity  Concentric strength testing of the rotator cuff muscles was completed on a Cybex II Dynamometer. The subjects completed all the testing on the right side first, then tested  70  the left after a short rest period. Each test consisted of three maximal internal and external rotation contractions at ninety and one hundred and eighty degrees per second to determine peak strength. An endurance test consisted of maximal fifty contractions at two hundred and forty degrees per second. Several calculations were made to completely evaluate rotator cuff strength and endurance. Peak strength values were divided by body weight to calculate relative strength. The peak external rotation value was subtracted from the internal to calculate the difference in internal and external rotation strength. Finally the internal rotation value was divided by the external to determine an internal to external strength ratio. Rather than examining peak values from the endurance test, an endurance ratio was calculated. The internal endurance ratio was calculated by dividing the mean of the three weakest consecutive contractions of the last ten repetitions by the mean of the three strongest consecutive contractions of thefirstten repetitions. The same formula was used for the external endurance ratio. Total endurance was calculated by adding the three lowest internal and external values and dividing by the three highest internal and external numbers. All absolute values are presented in foot-pounds of torque. The MANOVA performed on the strength data confirmed a difference in strength between the three injury groups (p = 0.003). ANOVA analysis did not identify a significant difference on any of the strength scores. A significant difference was found for the history group to have reduced right external endurance (p = 0.003) and a trend towards significance for right total endurance (p = 0.066). Post-hoc testing revealed that the injured group was not significantly different from the history group for right external endurance (p = 0.083), right total endurance (p = 0.125), or left external endurance (p = 0.319).  71  Each athlete performed a series of wall push-ups both before and after completing rotator cuff strength testing. During this test each athlete was examined for the occurrence of scapular winging. Only two injured athletes (one masters athlete and one injured) and none of the non-injured or history group displayed any significant scapular winging.  Absolute left int @ 907sec Absolute left ext @ 90°/sec Relative left int @ 907sec Relative left ext @ 90°/sec Left IR -ERdiff.@ 90 7sec Left IR/ER ratio @ 90 7sec Absolute left int @ 1807sec Absolute left ext @ 1807sec Relative left int @ 1807sec Relative left ext @ 1807sec Left IR - ER diff. @180 7sec Left IR/ER ratio @180 7sec Absolute right int @ 907sec Absolute right ext @ 90 7sec Relative right int @ 907sec Relative right ext @ 907sec Right IR - ER diff. @ 90 7sec Right IR/ER ratio @ 90 7sec Absolute right int @ 1807sec Absolute right ext @ 1807sec Relative right int @ 1807sec Relative right ext @ 1807sec Right IR-ER diff. @ 1807sec Right IR/ER ratio (&180 7sec Table XXI. Mean±S.D.  Injured  Non-injured  History  P value  29.86±10.7 19.72+8.0 0.20+06 0.13+05 10.14+5.1 1.57+28 22.68+10.3 14.82+6.9 0.151.06 0.10+05 7.87+5.1 1.54+29 32.08+11.6 21.97+9.0 0.22+07 0.15+06 10.11+5.1 1.50+26 27.50+11.6 17.58±7.9 0.18+.07 0.12+05 9.93±7.2 1.67+51  28.54+8.7 19.08+7.5 0.20+.05 0.13+.04 9.46+5.4 1.56+35 24.13+9.3 15.0+6.4 0.17+05 0.10+04 9.13+4.7 1.68+47 30.79+10.1 19.62+6.3 0.21+06 0.14+04 11.17+5.9 1.60+.34 24.63+9.3 15.97±5.4 0.17+.06 0.11+03 8.66±4.9 1.53+23  26.27+10.0 17.19+7.7 0.17+05 0.11+04 9.07±4.3 1.60+37 22.22+9.2 14.13+6.8 0.15±.05 0.09+.04 8.09+4.1 1.66+.38 30.42±11.7 19.58+8.2 0.20+06 0.13+04 10.85+5.1 1.60+29 24.67+11.9 15.11+6.9 0.16+06 0.10+04 9.56+5.6 1.63+21  0. 187 0..187 0. 120 0 153 0. 589 0.961 0.803 0.604 0.720 0.622 0.985 0.706 0.441 0.210 0.627 0.281 0.922 0.338 0.365 0.249 0.612 0.212 0.293 0.098  age group rotator cuff internal and external strength data  72  Left internal endurance Right internal endurance Left External Endurance Right External Endurance Left Total Endurance Right Total Endurance Table XXII. Mean±S.D.  Injured  Non-injured  History  /; value  66.42±12.7 61.28+18.2 60.54+18.8 62.65+15.0 64.38+12.6 61.86+16.3  64.57+13.2 65.91+13.3 67.18 ±17.9 68.09+15.2 65.69+13.3 66.68+12.0  61.51+16.3 62.63+13.8 57.02+25.3 54.18+19.7 60.04+17.3 60.0+13.1  0.642 0.418 0.193 0.003 0.333 0.066  age group rotator cuff internal and external endurance data  The stretching patterns of the athletes were also examined with the questionnaire. Subjects answered questions on the age they began stretching, number of times per week they stretched and the average length of each stretching session in minutes. The MANOVA was non-significant for the stretching data (p = 0.172). A trend towards injured athletes beginning to stretch at a younger age was apparent from the data (p = 0.076).  Age began Times/week Length of time  Injured  Non-injured  History  /; value  8.98+2.71 6.68+2.94 15.48+9.48  9.80+3.08 7.45+2.34 13.75+6.63  10.81+2.69 7.10+3.24 12.73+5.31  0.076 0.604 0.293  Table XXIII. MeandS.D.  stretching data for age group swimmers  The influence of technical errors in the development of impingement syndrome was addressed with use of a second questionnaire. Swim coaches completed a questionnaire for each of their athletes that participated in the study. The questionnaire examined each swimmer's breathing pattern, body roll, elbow recovery height, and number of strokes per length. The data from the questionnaires was organized in the following manner. Coaches responded yes (1) or no (2) to the question of athletes breathing more frequently on one side compared to the other. The breathing side was scored as right (1), left (2), or both equally (3). The remaining questions used a five-point scale to evaluate each question. There was no significant difference in technique (p -  73  0.697) found from the MANOVA analysis. ANOVA also failed to identify differences for any measurement.  Breathe one side Which side breathe Body roll Right elbow recovery Left elbow recovery Number of strokes  Injured  Non-injured  History  /; value  1.36±.50 2.00+88 2.86+1.10 2.86+86 2.86±.86 3.14+86  1.54+.51 2.23+91 2.69+1.13 2.97+89 2.97±.92 3.40+. 81  1.50+51 2.15+94 2.79+85 3.24±.96 3.15+96 3.35+81  0.499 0.733 0.847 0.321 0.559 0.605  Table XXIV. MeandS.D.for  age group technique analysis  The masters athletes were examined in this study to determine if the etiology of shoulder impingement is the same in age group and masters athletes. To answer this question, the masters were compared to the non-injured age groupers. The data was then analyzed using MANOVAs and ANOVAs to see if similar trends existed in the injured age group and masters swimmers. The training volume and duration is significantly different between the two groups (p < 0.001). The combination of swimming strokes used during practice was not different however.  Dryland training Months/year DaysAveek Hours/day Meters/practice MetersAveek m. frontcrawlAveek % frontcrawl % backstroke % breaststroke % butterfly Table XXV. MeansdS.D.  Masters injured  Non-injured  age group  1.42+1.11 10.5+91 3.041.62 1.13+17 2504.551367.7 802711796 466012036 61.25%118.3% 13.94%19.7% 14.56%18.8% 10.19%17.9%  3.8311.38 10.5511.14 6.061.46 3.111.94 5442.311059.8 46765112377 28671112819 56.9%118.2% 19.3%115.2% 9.2%17.7% 14.0%19.0%  p value  < 0.001 0.619 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.470 0.333 0.065 0.174  training volume masters athletes  74  Differences between the groups were also detected for years of competitive swimming and current best times (p < 0.001). The masters swimmers had been training for more years but had slower current times. No difference in race distance was noted. Masters injured Years competitive swimming 14.08±11.0 100 meter freestyletime(sec) 72.66±7.51 200 meter freestyletime(sec) 160.44119.59 Race Distance (meters) 246.5+280.0  Non-injured  p value  7.61 ±2.3 60.30 + 5.1 132.26 ± 17.3 376.20 ±425.9  0.081 < 0.001 0.001 0.908  Table XXVI. MeansdS.D. years swimming, race distance, and current best times  The chronically injured masters athletes had significantly greater pain levels than the non-injured age-groupers (p < 0.001). Masters Pain history Current pain Pain rating  injured  12 (100%) 11 (92%) 3.88±1.76  Non-injured  /; value  29 (72.5%) 6(15%) 0.68±.81  0.044 < 0.001 < 0.001  Table XXVII. Masters current and past shoulder pain  The number of positive impingement tests was significantly greater in the masters athletes (p = 0.004). As noted in the injured age-group swimmers, the Neer test is more reliable for indicating injury than the Hawkins test. However, the results were not as accurate as the age group swimmers. The masters athletes reported six right side, four left side and two bilateral injuries. Positive responses were noted for the Neer test in eight (150%) of left injured shoulders and six (75%) right injured shoulders. (The percentage of positive responses is calculated by dividing the number of positive impingement tests by the number of injured shoulders). Nine (150%) left injured shoulders and ten (125%) right injured shoulders had positive responses from the Hawkins tests. A positive Speed's test was noted in three (50%) and two (25%) of the left and right injured shoulders.  75  Masters injured  Non-injured  /; value  6(15%) 10 (25%) 23 (57.5%) 23 (57.5%) 0 (0%) 0 (0%)  < 0.001 0.116 0.325 0.128 0.001 0.072  8 (67%) Left Impingement Test 6 (50%) Right Impingement Test 9 (75%) Left Hawkins Test 10 (83%) Right Hawkins Test 3 (25%) Left Speed's Test 2 (17%) Right Speed's Test Table XXVIII. Results of masters impingement and biceps resistance  tests.  The MANOVA for rotator cuff range of motion demonstrates a significant reduction of flexibility in the masters swimmers (p < 0.001). Significant reductions in active internal rotation and cross-chest adduction were noted in the masters. Average active elevation in the masters group was 180° for the left shoulder and 179.58° for the right. Painful elevation was noted in four left shoulders and three right shoulders. Pain was first noted in the left shoulder at between 90° and 175° with a mean of 128.75°. Right shoulder pain was noted between 60° and 180° with a mean of 136.67°.  Left passive int rotation Right passive int rotation Left passive ext rotation @90° Right passive ext rot @90° Left passive ext rotation @0° Right passive ext rot @0° Left active int rotation Right active int rotation Left cross-chest adduction Right cross-chest adduction Table XXIX. Means±S.D.  Masters injured  Non-injured  p value  105.67°±23.46° 107.92°±21.61° 109.75°±19.59° 115.08°±19.89° 104.75°±20.30° 106.17°±19.35° 14.83±2.25 13.08±1.78 11.33 cm±4.83 14.0 cm+5.29  107.9°±19.9 107.9°±20.7 115.8°±12.2 118.7°±14.3 100.5°±16.2 99.5°±15.7 15.8±1.3 15.4±1.9 2.6cm±2.8 3.7 cm+3.3  0.744 0.962 0.223 0.540 0.490 0.242 0.063 < 0.001 < 0.001 < 0.001  for masters flexibility data  The non-injured athletes had significantly less glenohumeral joint laxity than the masters (p = 0.002). Excess laxity was noted on anterior drawer and inferior sulcus in the masters group. The non-injured group had a much greater incidence of generalized joint laxity.  76  Left anterior drawer Right anterior drawer Left posterior drawer Right posterior drawer Left inferior sulcus Right inferior sulcus Generalized Joint Laxity Generalized Joint Laxity 2  Musters injured  Non-injured  p value  0 29+54 0 88+43 0 63+83 1 04+62 1 42+63 1 17+62 1 (8%) 43.33+19.35  0.33+47 0.35+48 0.68+69 0.65+70 0.83+40 0.75+45 18 (45%) 20.88+24.9  0.921 0.002 0.864 0.103 < 0.001 0.016 0.025 0.007  Table XXX. MeansdS.D. for masters joint laxity Masters injured Absolute left int @ 907sec Absolute left ext @ 907sec Relative left int @ 907sec Relative left ext @ 90 7sec Left LR - ER diff @ 90 7sec Left IR/ER ratio @ 90 7sec Absolute left int @ 1807sec Absolute left ext @ 1807sec Relative left int @ 1807sec Relative left ext @ 1807sec Left IR - ER diff @180 7sec Left IR/ER ratio @180 7sec Absolute right int @ 907sec Absolute right ext @ 907sec Relative right int @ 907sec Relative right ext @ 907sec Right IR - ER diff @ 90 7sec Right IR/ER ratio @ 90 7sec Absolute right int @ 1807sec Absolute right ext @ 1807sec Relative right int @ 1807sec Relative right ext @ 1807sec Right IR-ER diff @ 1807sec Right IR/ER ratio @ 180 7sec Table XXXI. MeandS.D.  29.09+10.0 17.14+5.02 0.17+03 0.10+03 11.96+7.0 1.71+36 22.23+7.4 11.77+2.9 0.13+03 .07+02 10.46+6.53 1.91+54 31.55+9.97 17.91+6.3 .19+05 .11+03 13.64+5.3 1.83+34 24.46+6.5 13.75+6.1 .15+03 .08+03 10.71+4.26 2.00+73  Non-injure d  28.54+8.7 19.08+7.5 0.20+05 0.13+.04 9.46+5.4 1.56+35 24.13+9.3 15.0+6.4 0.17+05 0.10+04 9.13+4.7 1.68+47 30.79+10.1 19.62+6.3 0.21+06 0.14+04 11.17+5.9 1.60+34 24.63+9.3 15.97+5.4 0.17+06 0.11+03 8.66+4.9 1.53+23  p value  0.270 0.800 0.440 0.156 0.120 0.230 0.672 0.382 0.254 0.059 0.128 0.022 0.540 0.779 0.241 0.050 0.148 0.081 0.793 0.593 0.235 0.045 0.271 0.017  masters rotator cuff internal and external strength data  Shoulder strength and endurance in the masters rotator cuff was significantly lower than the non-injured (p = 0.009). A trend towards reduced external rotation and  greater internal to external rotation ratio is noted in the masters group. All of the endurance tests displayed a trend or were significantly lower in the masters group.  Left internal endurance Right internal endurance Left External Endurance Right External Endurance Left Total Endurance Right Total Endurance Table XXXII. Mean±S.D.  Masters injured  Non-injured  /; value  51.37+16.6 56.62+15.6 39.37+25.4 57.99+20.2 48.36+15.9 57.55+14.0  64.57+13.2 65.91+13.3 67.18+17.9 68.09+15.2 65.69+13.3 66.68+12.0  0.065 0.065 0.015 0.095 0.024 0.051  masters rotator cuff internal and external endurance data  Differences in stretching patterns were also noted between the two groups (p < 0.001) The masters athletes stretched fewer days and for less time per session than the controls.  Times/week Length of time  Masters Injured  Non-injured  p value  4.88+1.51 11.14+4.52  7.45+2.34 13.75+6.63  0.001 0.101  Table XXXIII. MeandS.D.  stretching data for masters swimmers  Differences in technique were noted between the masters and non-injured athletes, (p = 0.035) The masters athletes swam freestyle with significantly more body roll (p = 0.036) and less strokes per length (p < 0.001).  Breathe one side Which side breathe Body roll Right elbow recovery Left elbow recovery Number of strokes Table XXXIV. MeaniS.D.  Masters injured  Non-injured  p value  1.714+49 2.43+98 3.64+48 3.43+1.13 3.57+79 2.14+69  1.54+51 2.23+91 2.69+1.13 2.97+89 2.97+92 3.40+81  0.382 0.565 0.036 0.249 0.122 < 0.001  for masters technique analysis  The effect of acromial morphology on the development of rotator cuff injury was examined in the injured age-group and masters athletes. Each subject was given a series of x-rays including a true Anterior-Posterior view, an Axillary view, and a Supraspinatus  78  Outlet View radiograph. The shape of the acromion was classified, from the Outlet view as either type I (flat), II (curved), or IU (hooked) according to Bigliani's protocol. Results are presented in table XXXV. All of the radiographs were inspected by one radiologist. No difference was noted between the age-group and masters athletes in acromial morphology. A type I acromion was identified in 40% of the age-group and 44% of the masters. A type II was seen in 50% of the age-group and 44% of the masters. Finally, a type III acromion was identified in 10% of the age-group and 11% of the masters athletes. The acromioclavicular joint was also examined for the formation of osteophytes. None of the radiographs displayed any evidence of osteophyte formation. Participation in the x-ray portion of the study was voluntary and several subjects chose not to have them taken. Additionally, several subjects did not have the complete series of radiographs taken and could not be examined. Consequently, only nineteen shoulders radiographs were examined. Type I 4 (40%) 4 (44%) 8 (42%)  Type II 5 (50%) 4 (44%) 9 (47%)  Type III 1 (10%) 1 (11%) 2(11%)  Injured age-group Injured masters Total Table XXXV. Acromial morphology of Injured age group and Masters swimmers  Questionnaire data was collected on each athlete's dryland training routine and activities, weight training activities, and stretching patterns. The results of this questionnaire data are presented in the following graphs. Over ninety percent of all age group swimmers performed some type of dryland training. Running and abdominal exercises were the most commonly practiced activities.  79  100.00% i 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 20.00% 10.00% 0.00%  H injured • non-injured ED history • masters  Figure 4. Percentage participation  in dryland training activities by injury category  Almost one hundred percent of the age group swimmers practiced some form of resistance training (weights). Free weight use is by far the most common however many also used nautilus equipment, swim benches and surgical tubing for resistance training.  weights  free weights  Figure 5. Percentage participation  nautilus  swmbench surgical tube  other  in weight training activity by injury status  80  One hundred percent of the athletes in this study stretched on a regular basis. Shoulders were the most commonly stretched muscle group, especially among the injured and masters athletes. Stretching of the quadriceps and hamstrings were also regularly stretched.  Figure 6. Percentage stretching of each muscle group by injury status  Competitive swimmers are prone to develop other injuries besides rotator cuff tendinitis. Although shoulder injuries are by far the most common, many others may develop especially in the knees and back. The subjects in this study reported all injuries that occurred as a result of swimming. Both overuse and traumatic injuries are listed by area and percent in figure 7.  81  intercostal  Figure 7. Total swimming related injuries  Logistic regression analysis was completed on the data. The regression analysis examined differences between the injured and non-injured athletes. The forward stepwise method was used to calculate the variables. The results of the analysis are presented in table XXXVI. The Goodness of Fit for the analysis was 65.241. The -2 Log Likelihood was 61.277. No interaction effect was found between any variables. The exp. (B) value estimates the increased risk associated with a single unit change in the associated variable. For example, a one centimeter decrease in cross-chest adduction translates to a 1.49 times greater likelihood of injury.  82  Variable  li  Std. Error Significance Exp. (li)  L cross chest adduction L anterior drawer L sulcus L Neer impingement Constant  0.3994 1.1184 1.8029 2.3504 -6.5399  0.0985 0.5473 0.6897 0.6678 1.3168  1.4909 3.0601 6.0674 10.4896  0.0001 0.0410 0.0090 0.0004 0.0000  Table XXXVI. Results of Logistic Regression Analysis  The results of the logistic regression analysis can be used to create a formula based on the standard equation: Log-odds = a + bixi  + b2X2 + b 3 X  3  +  biX;.  This equation  was used to derive the following equation: Log-odds = -6.5399 + .3994(left cross chest adduction) + 1.1184(left anterior drawer) + 1.8029(left sulcus) + 2.3504(left Neer impingement) Other variables were analyzed with the logistic regression however adding additional variables did not increase the statistical significance of the equation. The same formula could be derived from the right side data but the left side has a slightly greater statistical significance.  83  V. DISCUSSION Injuries in almost any competitive sport are not uncommon. Swimmers are certainly no exception to the rule. Swimming is most often associated with over-use shoulder injuries. However, there are several other injuries that are very frequent in competitive swimming. The subjects were questioned on any injuries they developed as a result of swimming. Shoulder injuries were by far the most common, comprising sixty four percent of all injuries. Knee and back problems were the second and third most common injury at fifteen and eight percent respectively. Most of the injuries were overuse but a few traumatic injuries were reported to the head, ankles, and fingers. Most of these occur from hitting the wall while turning. The overhead athlete's shoulder is subjected to forces that the human shoulder was never intended to cope with. Consequently, shoulder injury is very common in this group of athletes. The testing of one hundred and seven competitive swimmers provided relevant information regarding the etiology of shoulder impingement syndrome and the condition of injured and non-injured athlete's shoulders. A combination of impingement tests and questionnaires were used to evaluate the degree of pain and dysfunction in the rotator cuff muscles. As the casual observer would probably assume, the injured group had a significantly greater history of pain (p < 0.001), occurrence of current pain (p < 0.001), and current pain rating (p < 0.001). The chronically injured masters group also had significantly greater results in all three measurements. The injured group also had more positive impingement tests (p < 0.001). The Neer impingement test was more reliable at diagnosing injury accurately. The test was  84  positive for eighty-nine percent of the injured left shoulders and 100% of the right. False positives were noted in the non-injured group atfifteenpercent and twenty-five percent for the left and right respectively. The Hawkins test recorded positive results in 105% of the left injured shoulders and 142% of the right and 57.5% of both the left and right noninjured shoulders. The Neer impingement test was positive in 130% of the left and seventy-five percent of the right injured shoulders of the masters group. The Hawkins test was positive for 150% and 125% of the left and right respectively. The Speed's test was positive infiftypercent of the left and twenty-five percent of the right in the masters group and thirty-three percent of the left and twenty-eight percentrightof the age groupers injured shoulders. The results of the impingement testing indicate that shoulder impingement syndrome is isolated to the supraspinatus muscle in most cases for both the age-group and masters swimmers. Approximately, only twenty-five tofiftypercent of athletes develop biceps tendinitis from competitive swimming. Significant differences were expected between the injured and non-injured subjects in pain ratings and impingement tests. The unexpected finding was the large number of non-injured and history of injury subjects that had positive impingement tests. 72.5% of the non-injured and 100% of the history group reported having shoulder pain at some point during their swimming career. Fifteen percent of the non-injured and twenty percent of the history group had shoulder pain at the time of examination. Positive impingement tests were noted in between fifteen and eighty percent of these athletes. A positive biceps resistance tested was found in 8.6% of the history group.  85  This series of testing suggests that rotator cuff injury in this population is not isolated to a specific sub-population but is experienced by almost all competitive swimmers. Ninety percent of the subjects in this study have had shoulder pain resulting from swimming. This finding emphasizes the difficulty of assigning subjects to a particular diagnosis category. All competitive swimmers probably have some degree of rotator cuff injury at certain points through their career. The difference between classifying an athlete as injured or non-injured depends on the severity of symptoms. If shoulder pain adversely affects competition or practice, that athlete would be considered injured. However, a swimmer with less severe pain or a higher pain threshold may be able to continue competing. Shoulder impingement syndrome should be viewed in competitive swimmers as a continuum rather than the presence or absence of injury. Most swimmers will develop some degree of shoulder pain during their career. The difference between the chronically injured and the successful athlete might be their ability to control shoulder injury. Therefore, all swimmers should take appropriate measures to prevent and/or treat injury in the rotator cuff. It is interesting to note that much of this testing was conducted during Christmas or spring training camps. The camps are usually associated with a rapid increase in training volume and duration. Consequently, several 'non-injured' athletes mentioned a recent occurrence of shoulder pain as a result of the change in training. This may partially explain the incidence of positive impingement tests and current shoulder pain among the non-injured athletes.  86  The etiology of shoulder impingement syndrome in the overhead athlete has been the subject of many studies since the 1970s. Many different etiologic factors and combination of factors have been suggested in the development of rotator cuff injury. To date however, no definitive explanation has been developed. This study investigated the role of seven common etiologic factors in the development of shoulder impingement syndrome in age group and masters swimmers. Differences in training volume and patterns; stroke technique; glenohumeral joint laxity; dysfunction of the scapulothoracic muscles; posterior capsule inflexibility; rotator cuff muscle strength and endurance; and acromial morphology were all thoroughly examined. There have been several research studies investigating proper swimming mechanics and the adverse effects of improper technique. Several common technical errors have been suggested in the etiology of rotator cuff injury. First, swimmers that breathe consistently to one side are at increased risk of injury. ' 30  74  Second, swimmers  with lower arm recovery tend to have more shoulder problems. Finally, athletes with 30  less body roll have an increased incidence of shoulder p a i n .  86  Testing in this study failed to find a significant difference (p = 0.697) in technique between the injury groups. Two possible conclusions could be drawn from this result. First, technical errors do not influence shoulder injury. Very few athletes have perfect technique. Perhaps, small idiosyncrasies in technique are not severe enough to negatively effect the rotator cuff muscle. The second possibility is that the test procedure was not sensitive enough to distinguish differences between the groups. The questions on the questionnaire used either a two or five point rating scale (see appendix B). These scales may have been  87  insensitive to minor differences in stroke. Additionally, approximately fifteen different swim coaches completed the questionnaires. Most of the questions were very subjective. Individual swim coaches may have different ideas as to what constitutes adequate body roll or elbow recovery. Therefore, variation in each coach's opinion of proper technique maybe reflected in the results of the questionnaire. The functioning of the scapulothoracic muscles was tested by having each subject complete a wall push-up. As they completed the push-up, the scapula was observed and assessed for winging. The presence of scapular winging is assumed to result from dysfunction of the scapulothoracic muscles. Scapular winging is typically attributed to weakness of the serratus anterior. However in swimmers, relative weakness of the rhomboids or imbalances between strong latissimus dorsi and weaker trapezius may also result in winging. Many authors ascribe the development of rotator cuff injury, at least partially, to improper functioning of these muscles. >>> - > > > However, this study 2  7  8  14 24  52  67  92  failed to consistently find scapular winging in the injured subjects. Only two of the subjects with shoulder impingement (one injured and one masters) displayed true scapular winging. The failure to find scapular winging in the current study could be the result of either insensitivity of the test protocol or gross scapular winging does not occur consistently in injured swimmers. Mild to moderate winging was noted in some athletes however gross scapular winging was required to be considered significant. Blevins hypothesized that "mild to moderate winging of the scapula is most commonly associated with scapular stabilizer asynchrony and dysfunction resulting from chronic shoulder pain." Based on information from the literature review and limited results of the study, 14  88  it appears that weakness or imbalance of the upper back musculature may influence shoulder injury. However, further research is required to completely assess the role of scapulothoracic dysfunction in the development of rotator cuff tendinitis. The development of rotator cuff tendinitis has been attributed to an acromion that intrudes into the subacromial space in several research papers. A prominent anterior acromion is one of the most commonly reported etiologic factors. ' ' ' ' ' ' ' ' 6  19  32  41,47  52  67  70  113  116  Several studies have investigated the incidence of each acromial type. Bigliani's early cadaver research reported rates of 17.1% for type I, 42.9% for type II and 39.3% for type III. A subsequent radiology study of patients with shoulder complaints reported incidences of 18%, 41% and 41% for type I, U and III respectively. Bigliani and other researchers claim these rates are consistent among age groups. ' 13  76  Several authors attribute shoulder impingement to a type III acromion  1 4 9 5  while  others claim both a type II and III influence injury. Bigliani found that 70% of cadavers 28  with a type III acromions had rotator cuff tears while only 3% of those with type I had tears. The formation of osteophytes on the anterior and inferior aspect of the acromion 13  can also cause inflammation and degeneration of the rotator cuff. ' '  6 66 80  Although only nineteen shoulders were examined, the data from the radiographs of the injured and masters subjects do not support these earlier findings. Forty percent of the injured age-group athletes were found to have a type I acromion, fifty percent a type II, and ten percent a type III. The masters athletes had rates of forty-four percent, fortyfour percent, and eleven percent for type I, II and III respectively. Combined rates for the two groups were forty two percent type I, forty seven percent type II and eleven percent type III. No evidence of osteophyte formation was noted for any subject.  89  There are a number of reasons that the results of this study can not be used to draw definite results. First, only nineteen radiographs were taken. This sample size is too small to generalize the results to the general population. Second, the radiographs were taken by several different technicians. Variations in the angle at which the x-rays were taken could effect the shape of the acromion on the radiograph Finally, several studies have shown that there is variability in the interpretation of the radiographs. Despite these limitations, there still several hypothesizes that can be made about the role of the acromion. First, it appears that a prominent anterior acromion is not a major influence in shoulder impingement syndrome in the young competitive swimmer. A type III acromion has typically been associated in the literature with rotator cuff injury. However, in this group of injured athletes, the incidence of a type III acromion is even lower than published averages presented for the entire population. Osteophytes, also commonly cited as a cause of injury, were also non-existent in the study group, even the older masters athletes. The low incidence of a type III acromion in this population could result for two reasons. First, a type III acromion might be age related and develop over time. The constant abrasion of the supraspinatus tendon on the undersurface of the acromion may cause alteration and deformation of the acromion. This would explain the high incidence of type III acromions in the cadaver studies. A second possibility is that individuals with a prominent type III acromion are incapable of competitive swimming for any length of time. If 70% of individuals with a type III acromion are likely to develop rotator cuff tears they will probably have shortened swimming career. An individual with a prominent anterior acromion would  90  also have an associated decrease in the subacromial space. This would translate to an increased amount of mechanical abrasion on the supraspinatus tendon. Consequently tendon degeneration could be accelerated. There could be many athletes that develop chronic shoulder injury and are unable to continue competitive swimming. Perhaps the incidence of type III acromion is more common in these athletes. Shoulder dysfunction in the young competitive swimmer might be a consequence not of the shape of the acromion, but the volume of structures passing beneath it. Greipp found a relationship between heavy weight lifting and shoulder injury in male swimmers. This, he hypothesized, is caused by hypertrophy of the supraspinatus muscle 39  and tendon. Possibly, many of the injured athletes have an adequate subacromial space. However, the demands of competitive swimming have resulted in larger muscles and tendons than a normal subacromial space can accommodate. Future research should investigate the relationship between tendon hypertrophy and shoulder injury in competitive swimmers. Overwork has been suggested by many authors as a prime cause of shoulder injury in the overhead athlete. > > - - > The repetitive nature of swimming can result in 2  17  32 42 52  97  rotator cuff fatigue • • • • * and also repeatedly forces the supraspinatus tendon between the head of the humerus and the coracoacromial arch. ' Although there is 53  87  much evidence indicating overuse in the development of shoulder impingement, no difference in workload was noted between the injured and non-injured age group swimmers. The MANOVA for the training data was non-significant (p = 0.631) as were the subsequent ANOVAs. Therefore, no differences were found in total duration of training, mileage, dryland training, or the percent each stroke was practiced. Richardson  91  and Miller noted a similar finding in 1991. They found that training distances were not related to the development of injury. Each athlete's competition level and race times 31  were also investigated. The hypothesis that faster swimmers, competing at higher levels would have to train longer and harder therefore would be more likely to develop shoulder injury was tested. Once again no difference was noted on any factor. Finally stroke specialization and race distance were examined. There has been some evidence to suggest that sprinters, especially those competing in freestyle or butterfly, were more susceptible to injury. ' 3 0  5 3 , 5 5  However, no significant difference in race specialization was detected  between the injury groups. Significant differences in training volume were noted between the masters and non-injured athletes. The masters athletes performed far less dryland training (p < 0.001) and practiced fewer days (p < 0.001) and hours per week (p< 0.001), and swam shorter distances (p < 0.001). Reduced training volumes in the masters group is a function of the training demands of masters swim clubs. These clubs in general train less frequently than the competitive age-group clubs. The results do not indicate that swimming less distance results in more shoulder injury. Although no significant difference was noted between the injured and non-injured groups on workload, competition level, or race specialization, overwork probably has a major role in athletes developing impingement syndrome. There is certainly an association between young athletes participating in overhead activities and the development of rotator cuff injury. The shoulder overwork involved in these sports combined with other anatomical and biomechanical factors probably results in rotator  92  cuff injury. Overwork alone does not result in shoulder dysfunction, but in conjunction with other weaknesses plays a role in the development of rotator cuff injury. Shoulder inflexibility is one weakness that could influence the development of shoulder impingement syndrome. Tightness in the posterior capsule of the shoulder has been suggested in the etiology of impingement. ' ' Posterior capsule inflexibility 14 66 110  results in superior translation of the humeral head. ' 30  110  This in turn increases the amount  of compression on the supraspinatus tendon between the humeral head and coracoacromial arch. Several investigators have also noted that injured athletes tend to have reduced internal rotation and cross-chest adduction. ' ' Greipp found a 8 11 110  correlation between anterior shoulder inflexibility and shoulder pain. No differences 39  have been found in external rotation or abduction. The decrease in cross-chest adduction 8  and internal rotation is thought to be caused by reactive fibrosis tissue in the capsule as a result of repetitive microtrauma. '  15 110  Evidence of posterior capsule inflexibility was apparent within the studied injured athletes. The injured age groupers displayed a trend towards reduction in passive right (p = 0.101) internal rotation. A more distinct trend is noted for limited left (p = 0.124) and right (p = 0.019) active internal rotation. Cross chest adduction is definitely reduced in the injured age-group athletes as compared to the non-injured athletes (left p < 0.001, right p - 0.042) Post-hoc testing reveals that the history group also trend towards posterior capsule inflexibility. The restricted range of motion on the posterior capsule is even more apparent in the masters swimmers. No difference was noted on passive internal rotation, but significant limitations were found for active internal rotation (left p  93  = 0.063, rightp < 0.001) and cross-chest adduction (leftp < 0.001, rightp < 0.001). No differences were noted in external rotation. No significant differences were noted in active abduction of the shoulder between injury groups. However, a few injured athletes had slightly reduced abduction. The classic painful arc of motion was not found in this population. A common finding in people with shoulder impingement syndrome is pain during 60° to 120° of shoulder abduction.  19,24  '  52  Painful abduction if present, was noted in the injured athletes beginning  between the range of 160° to 180°. The classic painful arc was not noted in this population because reduced or painful abduction is incompatible with competitive swimming. Most swimming strokes require excellent shoulder range of motion. Limitations of shoulder range of motion would prevent athletes from swimming competitively. As discussed earlier, the results of flexibility testing indicate a definite reduction in the flexibility of the posterior capsule in injured athletes. The superior migration of the humeral head, resulting from inflexibility, undoubtedly contributes to the development of shoulder impingement syndrome. Superior migration of the humeral head can result from inflexibility of the posterior capsule but may also be caused by the combination glenohumeral joint laxity and rotator cuff muscle imbalance. The rotator cuff muscles and the glenohumeral ligaments are the two structures primarily responsible for maintaining shoulder joint stability. Impairment in one structure often leads to dysfunction of the other resulting in glenohumeral joint instability.  Decrement in rotator cuff strength, especially the balance between internal and external rotation strength, has often been cited in the etiology of shoulder impingement. '  31 89  Swimming causes several problems within the rotator cuff muscles.  First, the freestyle and butterfly strokes selectively train the internal rotators. ' ' '  31 32 67 89  Unless corrective measures are taken, large muscle imbalances develop between the internal and external rotators. Second, swimming long distances fatigues the rotator cuff muscles. ' '  1 32 55  The fatigued rotator cuff muscles are very inefficient at maintaining  stability of the glenohumeral joint. Once the rotator cuff muscles are compromised, from either fatigue, muscle imbalance or a combination of the two, upward translation of the 1 0  O A O C SO  humeral head is inevitable. ' ' ' ' Excessive translation of the humeral head overloads the glenohumeral ligaments. If these forces act on the ligaments over a long period of time, these athletes 16  will develop laxity within the shoulder ligaments. Many authors have stated that swimmers rely heavily on the rotator cuff muscles because of acquired laxity in the glenohumeral joint. > > > > ' ' > A destructive cycle develops in the swimmers 2  6  14  ,7  32 52 55  60  shoulder where rotator cuff fatigue and imbalance overstress the glenohumeral ligaments. The laxity of the ligaments then force the cuff muscles to work harder to minimize shoulder translation and the cycle begins again. The results of this study provide evidence of this mechanism. No significant difference was discovered in absolute or relative internal and external strength, the difference between internal and external strength, or internal rotation to external rotation ratio in the age group athletes. The results indicate that all swimmers have muscle imbalances between the internal and external rotators. The internal to external rotation  95  strength ratios range between 150% to 168% in the age group swimmers. No significant difference was noted on any measurement. The masters athletes had ratios in the range of 171% to 200%. The master's ratios were significantly greater than the non-injured for both left (p = 0.022) and right (p = 0.017) internal to external ratios at 180 degrees per second. Trends were noted in the masters subjects towards lower relative left external rotation strength at one hundred eighty degrees per second (p = 0.059), relative right external rotation at ninety degrees per second (p = 0.05), and relative right external rotation at one hundred eighty degrees per second (p = 0.04). All of the age group swimmers display similar imbalances in the rotator cuff muscles. The masters group displayed even more imbalance in the cuff muscles. Especially at the higher speeds (180 degrees per second) where it is more difficult to develop torque. These results indicate that limitations in external rotation strength maybe related to the development of rotator cuff injury. Significant differences were noted in the endurance ratios of the external rotators of the injured and history of injury groups (right p = 0.003). A trend towards reduced total endurance was also noted in these groups (right p = 0.066). The injured athletes had better (although not statistically significant) endurance ratios than the history group. This could be the result of injured athletes spending more time on strengthening the rotator cuff as part of a rehabilitative program. Only anecdotal evidence is available, but conversation with the injured athletes indicates that most of them are attending physiotherapy and actively strengthening the rotator cuff muscles. The emphasis injured athletes have placed on rehabilitating the shoulder muscles could have resulted in endurance values that are greater than the pre-rehab level.  96  Differences were noted only on the right side. This is most likely a result of the test protocol. Each subject completed the test on the right side first. This was a difficult and strenuous test. Many subjects exerted less effort when completing the second test. Additionally, many of the injured athletes did not complete the left side test as maximum strength testing aggravated the injured rotator cuff muscles. Therefore, the mean for the left side endurance ratios for the injured athlete is probably lower than reported because many of the injured athletes did not complete the testing. The masters athletes had significantly reduced left external endurance (p = 0.015) and left total endurance (p = 0.015). Trends towards significance were noted for left internal endurance (p = 0.065),rightinternal endurance (p = 0.065),rightexternal endurance (p = 0.095), andrighttotal endurance (p = 0.051). The results indicate a trend towards reduced endurance in athletes with shoulder impingement syndrome. These results reflect earlier research findings. Warner et al. found that internal to external strength ratios could approach 200% in athletes with shoulder impingement.  110  This was thought to represent a weakness of the external rotators. Beach et al. found 110  no significant correlation between external to internal rotation and abduction to adduction strength ratios.  1 1  They did find that as external endurance ratios decreased, the reported  level of pain and dysfunction increased.  11  The strength testing reveals two important findings. First, all swimmers exhibit some degree of muscle imbalance in the rotator cuff muscles. Therefore, rotator cuff muscle imbalance alone does not result in shoulder impingement syndrome. Second, injured athletes tend to have reduced endurance of the rotator cuff muscles, especially the external rotators. Even if an athlete has both muscle imbalance and low endurance in the  97  external rotators, shoulder impingement syndrome will probably not result. However, when combined with glenohumeral joint laxity rotator cuff tendinitis seems unavoidable. Laxity testing was performed in the anterior, posterior, and inferior directions. Examination of the age-group subjects discovered the injured group had significantly greater translation in anterior drawer (left p = 0.003, right p = 0.012) and inferior sulcus (left p = 0.007, right p < 0.001)than the non-injured group. Posterior drawer was not different between the groups. The history group was expected to have similar laxity findings as the injured group as ligaments do not shorten quickly. The history group had significantly greater laxity than the non-injured group in right anterior drawer and right inferior sulcus, and a trend towards greater left inferior sulcus. The masters athletes were found to have a trend towards more anterior drawer (left p = 0.921, rightp = 0.002) and significantly more inferior translation (left p < 0.001, right p = 0.016). These findings indicate general laxity of the Inferior Glenohumeral •s  Ligament Complex and the Superior Glenohumeral Ligament. The results provide further support for the Circle Concept of Instability. The odd results for anterior drawer in the masters group could be a result of the insensitivity of the grading scale and small group size. A value of one (1) was applied to a shoulder with more anterior drawer than the contralateral side. It appears that a score of one was more common on the right side than the left. While this value could represent a small difference in anterior translation, statistically the difference can be quite large. Additionally, with only twelve subjects, differences between the left and right side did not average out.  98  A relationship between laxity of the glenohumeral ligaments and shoulder dysfunction has been noted in several other studies. ' ' '  Zemek and Magee found  that elite swimmers have greater translation in the anterior and inferior direction. difference was noted in posterior laxity.  1 1 9  No  31,119  There has been some debate in the literature about the development of ligament laxity. Most authors believe that joint laxity is acquired, ' ' > > > however there is 2 14 16  some evidence for an inherent basis. ' 14  119  66  83  119  This study found no difference in generalized  joint laxity within the age group subjects. The masters athletes had significantly less generalized joint laxity (p = 0.025) but significantly more anterior and inferior translation. This finding implies that glenohumeral laxity is acquiredfromthe shoulder overuse associated with competitive swimming. Acquired laxity could resultfromoverstretching, incorrect stretching techniques such as ballistic stretching, or over-stressing the ligaments to maintain shoulder stability during swimming. There was also a trend in this study for the injured athletes to begin stretching at a younger age (p = 0.076). Perhaps, beginning to stretch at a young age overstresses ligaments before they develop fully. Further research is required to determine the effects of stretching on the developing ligaments and tendons. Then adequate guidelines can be established for stretching programs for the young swimmer. The results of the strength and laxity testing provide evidence for the theory that a combination of glenohumeral joint laxity, rotator cuff muscle imbalances and low endurance in the external rotators have a major influence in the development of shoulder impingement syndrome. The glenohumeral ligaments and rotator cuff muscles function together to maintain shoulder stability. However, a vast majority of athletes in this study  99  had very strong internal and relatively weak external rotators. The injured subjects also had decreased external rotation endurance and lax ligaments. Dysfunction of both the shoulder ligaments and rotator cuff muscles will result in an inherently unstable shoulder. Shoulder instability typically precedes superior migration of the humeral head. Superior migration increases the degree of mechanical impingement on the supraspinatus tendon. This typically causes tendon inflammation, degeneration, and eventually rupture. Therefore, it appears likely that joint laxity and rotator cuff dysfunction are prime factors in the etiology of shoulder impingement syndrome. The results of the regression analysis indicate that posterior capsule inflexibility and glenohumeral joint laxity are the two main factors in the development of shoulder impingement. With a one-centimeter decrease in cross chest adduction, an athlete is 1.49 times more likely to develop rotator cuff injury. A one unit increase in anterior drawer translates to a 3.06 times greater chance of developing injury. A one-centimeter increase in inferior sulcus equals a 6.07 times greater possibility of shoulder injury. Finally, athletes with a positive impingement test are 10.49 times more likely to be injured than those with a negative impingement. The formula derived from the regression analysis can useful in estimating the probability of an athlete developing shoulder injury. The regression analysis provides further evidence for posterior capsule inflexibility and glenohumeral joint laxity as prime factors in the development of shoulder impingement syndrome. Although not a significant factor in the regression analysis, rotator cuff muscle imbalance also influences the formation of rotator cuff injury. Overwork certainly has an indirect role in the development of shoulder  100  impingement syndrome. Therefore, in the competitive swimmer, a combination of physiologic and biomechanical forces interact to cause rotator cuff injury.  101  CONCLUSIONS The etiology of shoulder impingement syndrome in the competitive swimmer has been studied for at least twenty-five years. To date however, no conclusive evidence has been published to explain the development of rotator cuff injury in the overhead athlete. This study investigated the role of seven common factors related to the development of shoulder injury. 1. There was no difference in workload between the injury categories of the age-group athletes. All competitive swimmers complete a very high workload. This should not be interpreted to mean that overwork does not cause shoulder injury. Large training volumes exaggerate other weakness within the shoulder. Workload then appears to be indirectly related to rotator cuff injury. 2. Distinct differences in glenohumeral laxity were noted in the injured swimmers. The injured subjects were found to have significantly greater shoulder translation in the anterior and inferior planes. While probably not sufficient to cause injury alone, combined with rotator cuff muscle imbalances and reduced external rotation endurance, joint laxity certainly influences rotator cuff degeneration. No difference in generalized joint laxity was noted. Therefore, excess laxity appears to be acquired from overwork and over-stretching. 3. Although there are several limitations in this portion of the study, the morphology of the acromion does not seem to cause rotator cuff injury in the young competitive swimmer. Previous research cited a type III acromion as a major contributing factor to rotator cuff injury. However, only 11% of the injured subjects in this study were found to have a type III acromion. The results suggest that changes in the shape of the  102  acromion maybe a natural age related degeneration and do not affect young individuals. A related hypothesis is that overhead athletes may develop tendon and muscle hypertrophy. Even in the athlete has a normal subacromial space there may not be enough space for the larger than average tendons. 4. Weakness or imbalance of the scapulothoracic muscles have been cited as a contributing factor in the development of shoulder injury. The protocol employed in this study however was insufficient to confirm scapulothoracic dysfunction. The lack of significant findings could be the result of insensitive test procedures or scapular winging may not occur frequently in injured swimmers. 5.  The injured athletes have a distinct limitation in the flexibility of the posterior capsule. Inflexibility of the posterior capsule causes anterior and superior migration of the humeral head. This naturally narrows the subacromial space and increases the degree of mechanical impingement on the supraspinatus tendon. The increased abrasion on the tendon will accelerate its degeneration and certainly has a influential role in the development of rotator cuff injury.  6. It was hypothesized before beginning this study that the injured athletes would have strength imbalances between the strong internal rotators and weaker external rotators. The study results indicate that not only the injured but also the non-injured swimmers display similar degrees of muscle imbalance. Both groups had internal rotation strength ranging from 150% to 200% of the external strength. The injured athletes do have a relative decrease in the endurance of the rotator cuff muscles, especially the external rotators.  103  7. Technical errors in swimming stroke were thought to increase the risk of shoulder injury. The measurement techniques used to evaluate stroke in this study did not find significant differences between the injured and non-injured groups. The lack of significant results could indicate that minor differences in technique are not a major influence in developing rotator cuff injury or that the test procedures were insensitive to differences between the groups. The results of the regression analysis emphasize the relative importance of posterior capsule inflexibility and glenohumeral joint laxity in the development of shoulder impingement syndrome. Swimmers with a decrease in cross chest adduction or increased anterior drawer and inferior sulcus are significantly more likely to develop rotator cuff injury. The results of this study certainly do not explain the development of shoulder impingement syndrome in all people. Elderly individuals almost certainly develop injury for different reasons than young athletes. Even among athletes, the etiology of rotator cuff injury is probably different in baseball players than swimmers. This study does give some good indication of the causes of shoulder injury in the competitive swimmer. Technical errors and acromial morphology so not seem to influence injury in the young swimmer. Imbalances of the rotator cuff muscles, weak external rotators, inflexibility of the posterior joint capsule, and acquired laxity of the glenohumeral ligaments combined with overwork seem to result in the development of shoulder impingement syndrome.  104  LIMITATIONS & FUTURE RESEARCH Several limitations exist within the design of this study. First, no assessment of acromial morphology is made in normal subjects. The shape and slope of the acromion can be determined from radiographs. However, it is not ethical to subject healthy individuals to the radiation associated with radiographs. If some non-injured subjects had radiographs taken of the shoulder, they would be examined to assess acromial morphology. However, none of the non-injured athletes have had shoulder x-rays taken. Therefore, the injured athletes can only be compared to normative results. Future research should assess the typical acromial morphology in the non-injured athlete and study the development of type II and III acromion and bone spurs over time. Injured athletes with a type I or II acromion could also be studied to determine if tendon and muscle hypertrophy have a significant influence in the development of shoulder impingement syndrome. A study utilizing Magnetic Resonance Imaging (MRI) would be useful for measuring the volume of structures in the subacromial space and any relationships to rotator cuff injury. Second, rotator cuff vascularity has not been examined. Much of the current literature still cites hypovascularity as a major factor in the development of rotator cuff injury. Unfortunately, there is not a non-invasive method of accurately assessing rotator cuff blood flow. Laser Doppler Flowmetery provides a very interesting possibility. However, with present technology, this device may only be used intra-operatively. Future research should focus on assessing the blood flow to the supraspinatus muscle during exercise and evaluating differences in flow volume between injured and non-injured shoulders.  105  Third, a thorough evaluation of the Serratus Anterior and upper back muscles should be made in competitive swimmers. Much current research suggests that weakness or muscle imbalance resulting in scapulothoracic dysfunction is a major factor in the development of rotator cuff injury. Using a wall push-up to test the functioning of these muscles was not sensitive enough to recognize differences between the injured and noninjured athletes. The examiner did notice differences in the movement of the scapula in some subjects. However none displayed true scapular winging and no method was employed to quantify slight variation between subjects. A further study using EMG technology to exhaustively appraise the functioning of the scapulothoracic muscles would useful at identifying weakness and imbalance in injured athletes. Fourth, the analysis of the swimming stroke technique in this study was too subjective and completed by too many people. A controlled assessment of swimming technique could be performed by one examiner using video analysis. This data could be used to address differences in the technique of injured and non-injured athletes. Finally, many of the shoulder laxity measurements have subjective scoring methods that are not sensitive enough to detect minor differences between subjects. For example, a value of one is given if movement in one shoulder is greater than the contralateral side and two is given is the shoulder can be subluxed over the glenoid rim. There can be a great deal of variation between an athlete with a slight amount of movement in the joint compared to an athlete whose shoulder can be moved to the glenoid rim but not sublux over it. Both of these subjects, however, would be given a score of one with the evaluation system used in this study. A more graduated measuring tool would be much more sensitive to judging differences in shoulder ligament laxity.  106  BIBLIOGRAPHY 1. Abrams JS. Special shoulder problems in the throwing athlete: pathology, diagnosis, and nonoperative management. Clinics in Sports Medicine 1991; 10(4):839-61. 2. Allegrucci M, Whitney SL, Irrgang JJ. Clinical implications of secondary impingement of the shoulder in freestyle swimmers. Journal of Sport and Physical Therapy 1994;20(6):307-18. 3. Andrews JR, Byrd JWT, Kupferman SP, Angelo RL. The profile view of the acromion. Clinical Orthopaedics and Related Research 1991;263:142-6. 4. Aoki M, Ishii S, Usui M. The slope of the acromion and rotator cuff impingement. Orthopaedic Transcripts 1986;10:228 5. Aoki M, Ishii S, Usui M, Post M, Morrey BF, Hawkins RJ, editors. Surgery of the shoulder. Toronto: Mosby Year Book; 1990; 46, Clinical application for measuring the slope of the acromion, p. 200-3. 6. Arroyo JS, Hershon SJ, Bigliani LU. Special considerations in the athletic throwing shoulder. Orthopedic Clinics of North America 1997;28(l):69-78. 7. Bak K, Fauno P. Clinical findings in competitive swimmers with shoulder pain. The American Journal of Sports Medicine 1997;25(2):254-60. 8. Bak K, Magnusson SP. Shoulder strength and range of motion in symptomatic and pain-free elite swimmers. The American Journal of Sports Medicine 1997;25(4):454-9. 9. Banas MP, Miller RJ, Tatterman S. Relationship between the lateral acromion angle and rotator cuff disease. Journal of Shoulder & Elbow Surgery 1995;4(6):454-61. 10. Bartlett LR, Storey MD, Simons BD. Measurement of upper extremity torque production and its relationship to throwing speed in the competitive athlete. The American Journal of Sports Medicine 1989;17(1):89-91. 11. Beach ML, Whitney SL, Dickoff-Hoffman S. Relationship of shoulder flexibility, strength, and endurance to shoulder pain in competitive swimmers. Journal of Sport and Physical Therapy 1992;16(6):262-8. 12. Bigliani LU, Morrison DS, April EW. The morphology of the acromion and its relationship to rotator cuff tears. Orthopaedic Transcripts 1986;10:228 13. Bigliani LU, Ticker JB, Flatow EL, Solowsky LJ, Mow VC. The relationship of acromial architecture to rotator cuff disease. Clinics in Sports Medicine 1991; 10(4):82337.  107  14. Blevins FT. Rotator cuff pathology in athletes. Sports Medicine 1997;24(3):205-20. 15. Boublik M , Hawkins RJ. Clinical examination of the shoulder complex. Journal of Sport and Physical Therapy 1993;18(l):379-85. 16. Bowen M K , Warren RF. Ligamentous control of shoulder stability based on selective cutting and static translation experiments. Clinics in Sports Medicine 1991;10(4):757-82. 17. Bowen M K , Warren RF, Harries M , Williams C, Stamish W, Micheli LJ, editors. Oxford textbook of Sports Medicine. New York: Oxford Medical Publications; 1994; 4.3.3, Injuries of the rotator cuff. p. 442-7. 18. Breazeale N M , Craig E V . Partial-thickness rotator cuff tears: pathogenesis and treatment. Orthopedic Clinics of North America 1997;28(2): 145-54. 19. Butters KP, Rockwood CA. Office evaluation and management of the shoulder impingement syndrome. Orthopedic Clinics of North America 1988;19(4):755-65. 20. Chansky H A Iannotti JP. The vascularity of the rotator cuff. Clinics in Sports Medicine 1991;10(4):807-21. 21. Ciullo JV. Swimmer's shoulder. Clinics in Sports Medicine 1986;5(1): 115-37. 22. Ciullo JV, Guise R. Adolescent swimmer's shoulder. Orthopaedic Transcripts 1983;7:171 23. Cone RO, Resnick D, Danzig L. Shoulder impingement syndrome: radiographic evaluation. Radiology 1984;150(l):29-33. 24. Corso G. Impingement relief test: an adjunctive procedure to traditional assessment of shoulder impingement syndrome. Journal of Sport and Physical Therapy 1995;22(5): 18392. 25. Davies GJ, Dickoff-Hoffman S. Neuromuscular testing and rehabilitation of the shoulder complex. Journal of Sport and Physical Therapy 1993;18(2):449-58. 26. Delee JC, Drez D, editors. Orthopaedic Sports Medicine: Principles and Practice. Toronto: W B Saunders Co. 1994; 15, The shoulder, p. 751-5. 27. Epstein RE, Schweitzer M E , Frieman BG, Fenlin JM, Mitchell D G . Hooked acromion: prevalence on M R images of painful shoulders. Radiology 1993;187(2):47981. 28. Fongemie A E , Buss DD, Rolnick SJ. Management of shoulder impingement syndrome and rotator cuff tears. American Family Physician 1998;57(4):667-74.  108  29. Foster CR. Multidirectional instability of the shoulder in the athlete. Clinics in Sports Medicine 1983;2(2):355-67. 30. Fowler PJ. Swimmers problems. The American Journal of Sports Medicine 1979;7(2): 141-2. 31. Fowler PJ, Webster MS. Shoulder pain in the highly competitive swimmer. Orthopaedic Transcripts 1983;7:170 32. Fowler PJ, Nicholas JA, Hershman EB, editors. The upper extremity in sports medicine. Toronto: C.V. Mosby Co. 1990; 40, Upper extremity swimming injuries, p. 891-6. 33. Frieman BG, Albert TJ, Fenlin JM. Rotator Cuff disease: a review of diagnosis, pathophysiology, and current trends in treatment. Archive of Physical Medicine and Rehabilitation 1994;75:604-9. 34. Frisiello S, Gazaille A, O'Halloran J, Palmer ML, Waugh D. Test-retest reliability of eccentric peak torque values for shoulder medial and lateral rotation using the biodex isokinetic dynamometer. Journal of Sport and Physical Therapy 1994; 19(6):341-4. 35. Gerber C, Scheenberger AG, Vihn T. The arterial vascularization of the humeral head. The Journal of Bone and Joint Surgery 1990;72-A(10): 1486-94. 36. Getz JD, Recht MP, Piraino DW, Schils JP, Latimer BM, Jellema LM, Obuchowski NA. Acromial morphology: relation to sex age, symmetry, and subacromial enthesophytes. Radiology 1996;199(3):737-42. 37. Giombini A, Rossi F, Pettrone F A Dragoni S. Posterosuperior glenoid rim impingement as a cause of shoulder pain in top level waterpolo players. The Journal of Sports Medicine and Physical Fitness 1997;37(4):273-8. 38. Grana WA, Teague B, King M, Reeves RB. An analysis of rotator cuff repair. The American Journal of Sports Medicine 1994;22(5):585-8. 39. Greipp JF. Swimmer's shoulder: the influence of flexibility and weight training. The Physician and Sportsmedicine 1985;13(8):92-105. 40. Hardy DC, Vogler III JB, White RH. The shoulder impingement syndrome: prevalence of radiographicfindingsand correlation with response to therapy. American Journal of Roentgenology 1986;147:557-61. 41. Hawkins RJ, Abrams JS. Impingement syndrome in the absence of rotator cuff tears (Stage 1 and 2). Orthopedic Clinics of North America 1987;18(3):373-82.  109  42. Hawkins RJ, Kennedy JC. Impingement syndrome in athletes. The American Journal of Sports Medicine 1980;8(3):151-8. 43. Hawkins RJ, Koppert G, Johnston G. Recurrent posterior instability (subluxation) of the shoulder. The Journal of Bone and Joint Surgery 1984;66-A(2): 169-74. 44. Hawkins RJ, Mohtadi NGH. Controversy in anterior shoulder instability. Clinical Orthopaedics and Related Research 1991;272:152-61. 45. Haygood TM, Langlotz CP, Kneeland BJ, Iannotti JP, Williams GR, Dalinka MK. Categorization of acromial shape: Interobserver variability with MR imaging and conventional radiology. American Journal of Roentgenology 1994;162:1377-82. 46. Hinton RY. Isokinetic evaluation of shoulder rotational strength in high school baseball pitchers. The American Journal of Sports Medicine 1988;16(3):274-9. 47. Hulstyn MJ, Fadale PD. Shoulder injuries in the athlete. Clinics in Sports Medicine 1997;16(4):663-79. 48. Jackson R, Fitch K, and O'Brien M, editors. Sports Medicine Manual. Calgary: Hurford Enterprises Ltd. 1990. 49. Jensen BR, Sjogaard G, Bommyr S, Arborelius M, Jorgensen K. Intramuscular laserdoppler flowmetery in the supraspinatus muscle during isometric contractions. European Journal of Applied Physiology 1995;71:373-8. 50. Jobe FW. Serious rotator cuff injuries. Clinics in Sports Medicine 1983;16(3):407-12. 51. Jobe FW, Radovich Moynes D. Delineation of diagnostic criteria and a rehabilitation program for rotator cuff injuries. The American Journal of Sports Medicine 1982;10(6):336-9. 52. Kamkar A, Irrgang JJ, Whitney SL. Nonoperative management of secondary shoulder impingement syndrome. Journal of Sport and Physical Therapy 1993; 17(5):212-24. 53. Kennedy JC, Hawkins RJ, Krissoff WB. Orthopaedic manifestations of swimming. The American Journal of Sports Medicine 1978;6(6):309-22. 54. Kitay GS, Iannotti JP, Williams GR, haygood TM, Kneeland BJ, Berlin J. Roentgenographic assessment of acromial morphologic condition in rotator cuff impingement syndrome. Journal of Shoulder & Elbow Surgery 1995;4(6):441-8. 55. Koehler SM, Thorson DC. Swimmer's shoulder: targeting treatment. The Physician and Sportsmedicine 1996;24(ll):39-50.  110  56. Leroux J, Codine P, Thomas E, Pocholle M, Mailhe D, Blotman F. Isokinetic evaluation of rotational strength in normal shoulders and shoulders with impingement syndrome. Clinical Orthopaedics and Related Research 1994;304:108-15. 57. Lichfield R, Hawkins RJ, Dillman CJ, Atkins J, Hagerman G. Rehabilitation for the overhead athlete. Journal of Sport and Physical Therapy 1993;18(2):433-41. 58. Ling S, Chen C, Wan R. A study of the vascular supply of the supraspinatus tendon. Surgical-Radiologic Anatomy 1990;12:161-5. 59. Lippitt S, Matsen F. Mechanisms of glenohumeral joint stability. Clinical Orthopaedics and Related Research 1993;291:20-8. 60. Lohr JF, Uhtoff HK. The microvascular pattern of the supraspinatus tendon. Clinical Orthopaedics and Related Research 1990;254:35-8. 61. MacKinnon P; Morris J. Oxford Textbook of Functional Anatomy. Toronto: Oxford University Press; 1994. 62. Malanga GA, Jenp YN, Growney ES, An KN. EMG analysis of shoulder positioning in testing and strengthening the supraspinatus. Medicine and Science in Sports and Exercise 1996;28(6):661-4. 63. McCoy RL, Dec KL, McKeag DB, Honing EW. Common injuries in the child or adolescent athlete. Primary Care 1995;22(1): 117-43. 64. McMaster WC, Troup J. A survey of interfering shoulder pain in United States competitive swimmers. The American Journal of Sports Medicine 1993;21(l):67-70. 65. Mehran JE, Vezina J. The etiology of chronic impingement syndrome reevaluated. Orthopaedic Transcripts 1987; 11:235 66. Meister K, Andrews JR. Classification and treatment of rotator cuff injuries in the overhand athlete. Journal of Sport and Physical Therapy 1993 ;18(2):413-20. 67. Miniaci A, Fowler PJ. Impingement in the athlete. Clinics in Sports Medicine 1993;12(1):91-109. 68. Morrison DS, Bigliani LU. The clinical significance of variations in acromial morphology. Orthopaedic Transcripts 1987;11:234 69. Mosely HF, Goldie I. The arterial pattern of the rotator cuff of the shoulder. The Journal of Bone and Joint Surgery 1963;45-B(4):780-9. 70. Neer CS. Anterior acromioplasty for the chronic impingement syndrome in the shoulder. The Journal of Bone and Joint Surgery 1972;54-A(l):41-50.  Ill  71. Neer CS. Impingement lesions. Clinical Orthopaedics and Related Research 1983;173:70-7. 72. Neer CS, Foster CR. Inferior capsular shift for involuntary inferior and multidirectional instability of the shoulder. The Journal of Bone and Joint Surgery 1980;62-A(3):897-907. 73. Neer CS, PoppenNK. Supraspinatus outlet. Orthopaedic Transcripts 1987;11:234 74. Neer CS, Welsh RP. The shoulder in sports. Orthopedic Clinics of North America 1977;8(3):583-91. 75. Neviaser RJ, Neviaser TJ. Observations on impingement. Clinical Orthopaedics and Related Research 1990;254:60-3. 76. Nicholson GP, Goodman DA, Flatow EL, Bigliani LU. The acromion: morphologic condition and age related changes. A study of 420 scapulas. Journal of Shoulder & Elbow Surgery 1996;5(1):1-11. 77. Nordgren B. Mechanical evaluation of muscle power in clinical practice with some aspects on training. Scandinavian Journal of Rehab Medicine 1994;Suppl 30:45-52. 78. Payne LZ, Deng XH, Craig EV, Torzilli PA, Warren RF. The combined dynamic and static contributions to subacromial impingement: a biomechanical analysis. The American Journal of Sports Medicine 1997;25(6):801-8. 79. Peh WCG, Farmer THR, Totty WG. Acromial arch shape: assessment with MR imaging. Radiology 1995;195(2):501-5. 80. Penny JN, Welsh RP. Shoulder impingement syndrome in athletes and their surgical management. The American Journal of Sports Medicine 1981 ;9(1): 11-5. 81. Perry J. Anatomy and biomechanics of the shoulder in throwing, swimming, gymnastics, and tennis. Clinics in Sports Medicine 1983;2(2):247-69. 82. Pink M, Jobe FW, Perry J, Kerrigan J, Browne A, Scovazzo ML. The normal shoulder during the butterfly stroke: An electromyographic and cinematographic analysis of twelve muscles. Clinical Orthopaedics and Related Research 1993;288:48-59. 83. Pollack BG, Bigliani LU, Flatow EL. The mechanical properties of the inferior glenohumeral ligament. New Orleans, LA: 1990; from "Ligamentous control of shoulder stability based on selective cutting and static translation experiments.". 84. Rathbun JR, MacNab I. The microvascular pattern of the rotator cuff. The Journal of Bone and Joint Surgery 1970;52-B(3):540-53.  112  85. Regan WD, Webster-Bogaert S, Hawkins RJ, Fowler PJ. Comparative functional analysis of the Brostow, Magnusson-Stack, and Putti-Platt procedures for recurrent dislocation of the shoulder. The American Journal of Sports Medicine 1989;17(l):42-7. 86. Renstrom PAFH, editors. Clinical Practice of Sports Injuries: Prevention and Care. Toronto: Blackwell Scientific Publications; 1994;Sport Specific Injuries, p. 508-9. 87. Richardson AB, Jobe FW, Collins HR. The shoulder in competitive swimming. The American Journal of Sports Medicine 1980;8(3): 159-63. 88. Richardson AB, Miller JW. Swimming and the older athlete. Clinics in Sports Medicine 1991;10(2):30T-11. 89. Richardson AR. The biomechanics of swimming: the shoulder and knees. Clinics in Sports Medicine 1986;5(1): 103-13. 90. Rockwood C A Lyons FR. Shoulder impingement syndrome: diagnosis, radiographic evaluation, and treatment with a modified Neer acromioplasty. The Journal of Bone and Joint Surgery 1993;75-A(3):409-23. 91. Rothman RH, Wesely WP. The vascular anatomy of the rotator cuff Clinical Orthopaedics and Related Research 1965;41:176-86. 92. Rupp S, Berninger K, Hopf T. Shoulder problems in high level swimmers impingement, anterior instability, muscular imbalance? International Journal of Sports Medicine 1995;16(8):557-62. 93. Sale DG. 1998; 3, Testing strength and power, p. 21-106. 94. Sarkar K, Taine W, Uhtoff HK. The ultrastructure of the coracoacromial ligament in patients with chronic impingement syndrome. Clinical Orthopaedics and Related Research 1990;254:49-54. 95. Scarpinato DF, Bramhall JP, Andrews JR. Arthroscopic management of the throwing athlete's shoulder: indications, techniques, and results. Clinics in Sports Medicine 1991;10(4):913-27. 96. Schwartz RE, O'Brien SJ, Warren RF, Torzilli PA. Capsular restraints to anteriorposterior motion of the shoulder. Orthopaedic Transcripts 1988;17:727 97. Silliman JF, Hawkins RJ. Current concepts and recent advances in the athlete's shoulder. Clinics in Sports Medicine 1991; 10(4):693-705. 98. Silliman JF, Hawkins RJ. Classification and physical diagnosis of instability of the shoulder. Clinical Orthopaedics and Related Research 1993;291:7-19.  113  99. Soslowsky LJ, Carpenter JE, Bucchieri JS, Flatow EL. Biomechanics of the rotator cuff. Orthopedic Clinics of North America 1997;28(1): 17-30. 100. Speer K. Anatomy and pathomechanics of shoulder instability. Clinics in Sports Medicine 1995;14(4):751-60. 101. Swain RA, Wilson FD, Harsha DM. The os acromiale: another cause of impingement. Medicine and Science in Sports and Exercise 1996;28(12): 1459-62. 102. Swiontkowski MF, Iannotti JP, Boulas HJ. Post M, Morrey BF, Hawkins RJ, editors. Surgery of the Shoulder. Toronto: Mosby Year Book; 1990; 48, Intraoperative assessment of rotator cuff vascularity using Laser Doppler flowmetery. p. 208-12. 103. Tibone JE, Jobe FW, Kerlan RK, Carter VS, Shields CL, Lombardo SJ, Yocum LA. Shoulder impingement syndrome in athletes treated by an anterior acromioplasty. Clinical Orthopaedics and Related Research 1985;198:134-40. 104. Ticker JB, Fealy S, Fu FH. Instability and impingement in the athlete's shoulder. Sports Medicine 1995;19(6):418-26. 105. Tuite MJ, Toivonen D A Orwin JF, Wright DH. Acromial angle on radiographs of the shoulder: correlation with the impingement syndrome and rotator cuff tears. American Journal of Roentgenology 1995;165(3):609-13. 106. Uhtoff HK, Sano H. Pathology of failure of the rotator cuff tendon. Orthopedic Clinics ofNorth America 1997;28(1):31-41. 107. Uhtoff HK, Sarkar K. An algorithm for shoulder pain caused by soft tissue disorders. Clinical Orthopaedics and Related Research 1990;254:121-7. 108. Van de Graaf KM. Human Anatomy. Dubuque: Wm. C. Brown Publishers; 1992. 109. Wallace WA; Hutson MA. Sports injuries: recognition and management. Toronto: Oxford Medical Publications; 1990. 33p. 110. Warner JP, Micheli LJ, Arstanian LE, Kennedy JC, Kennedy R. Patterns of flexibility, laxity, and strength in normal shoulders and shoulders with instability and impingement. The American Journal of Sports Medicine 1990; 14(4):366-75. 111. Warren RF, Kornblatt IB, Marchland R. Static factors affecting posterior shoulder stability. Orthopaedic Transcripts 1984;8:89 112. Watson M. Rotator cuff function in the impingement syndrome. The Journal of Bone and Joint Surgery 1989;71-B(3):61-6.  114  113. Watson M, Nicholas JA, Hershman EB, editors. The upper extremity in sports medicine. Toronto: C.V. Mosby Co. 1990; 8, Impingement and rotator cuff lesions, p. 213-9. 114. Wilk KE, Andrews JR, Arrigo C, Keirns MA, Erber DJ. The strength characteristics of internal and ext. rotator muscles in professional baseball pitchers. The American Journal of Sports Medicine 1993;21(l):61-6. 115. Wilk KE, Arrigo C. Current concepts in the rehabilitation of the athletic shoulder. Journal of Sport and Physical Therapy 1993;18(l):365-78. 116. Wolin PM, Tarbet JA. Rotator cuff injury: addressing overhead overuse. The Physician and Sportsmedicine 1997;25(6):55-74. 117. Yao L, Lee H, Gentili A, Shapiro MM. Lateral down-sloping of the acromion: a useful MR sign? Clinical Radiology 1996;51:869-72. 118. Yocum LA. Assessing the shoulder: history, physical examination, differential diagnosis and special tests used. Clinics in Sports Medicine 1983;2(2):281-9. 119. Zemek MJ, Magee DJ. Comparison of glenohumeral joint laxity in elite and recreational swimmers. Clinical journal of Sports Medicine 1996;6(l):40-7.  115  APPENDIX A - POWER STUDY The number of subjects required to find statistically significant results depends on the effect size or the magnitude of observed differences between groups. The effect size is standardized by taking the difference between sample means and dividing by their common standard deviation. This can be expressed in the following equation: d = X l JL.X2  s where d is the effect size, xi and x are group means, and s is their common standard 2  deviation. The value for d can then be used in a table to determine the number of subjects required. "It has been suggested that a power of 80% represents a reasonable protection against type II error. Some statisticians suggest that higher power should be attained, such as 90%." Both values will be given. 19  Studies by Warner et al. and Leroux et al. provide data on the strength and flexibility characteristics of people with shoulder impingement syndrome. Warner et al. internal to ext. rotation strength ratios of 200% in the impingement group and 135% in the normal group with a standard deviation of 35%. At an 80% power, n = 9, at 90% n = 12. Leroux et al. performed peak torque tests at 60° and 180° per second. Based On their results at 60° and 180°/second, 80% power n = 12, 90% power = 16. Based on a posterior drawer test performed by Leroux, n = 99 at 80% and 132 at 90%. The difference between groups on this factor was quite small. Leroux also measured range of motion in the shoulder. The normal group had a range of motion of 127.3° while the impingement group had 113° The 80% power tofinda statistical difference requires 26 subjects while 90% power needs 34 subjects.  116  Based on the results of these studies differences in internal and ext. rotational strength should be apparent with approximately 15 subjects. Finding a significant difference for range of motion and laxity may not as easy. At least 25 subjects will be required to determine the differences in range of motion.  117  A COMPREHENSIVE ANALYSIS OF THE SWIMMER'S SHOULDER QUESTIONNAIRE SUBJECT # Personal Information: 1. What is your age? 2. Are you male or female? 3. How much do you weigh? 4. Are you left or right handed? 5. How many years have you been swimming competitively? 6. What is the highest level you have competed at? (circle one) A) local  B) provincial  C) College/University  D)National  E) International  7. What are your current best times in the 100m and 200m freestyle? 100m  200m  8. Do you have a stroke and/or race distance specialization?  Yes  No  (If yes please list)  9. Have you ever had a swimming related injury?  '  Yes  No  10. Have you ever has shoulder pain while swimming?  Yes  No  11. Do you currently have shoulder pain while swimming?  Yes  No  (If yes please list)  (If yes, how long have you had shoulder pain?) 12. Please circle the number that corresponds to your present level of shoulder pain: 0 1 2 3 4 5 .6  no pain occasional shoulder pain which lasts less than two hours. No problem shoulder pain lasting longer than two hours following swim practice shoulder pain experienced on forceful arm movements shoulder pain which is annoying for perhaps hours a day. Could have affected my practice abilities. Pain was very annoying, Almost certainly affected my ability to practice hard Severe shoulder pain, lasting at least twelve hours per day (unless I used ice/medication etc.) Almost impossible to practice hard.  120  13. Have you ever been diagnosed with Shoulder Impingement Syndrome (rotator cuff injury?  Yes No  14. Have you ever had a shoulder injury from a sport other than swimming? Yes No (If yes, please list sport and injury)  Training Information: 15. How many months will you swim this year? 16. How many days per week do you swim? 17. How many hours per day do you swim (on average) 18. On average, how many meters do you swim per practice and week?  19. Approximately, how many meters of frontcrawl do you swim per week? 20. Approximately what percentage (%) of your workout is spent swimming each of the four major strokes?  21. How much time do you spend per week on dryland training? 22. What does your dryland training consist of?  23. Do you ever use weight training in your workouts?  Yes  No  24. Which type of weight training equipment do you use? (circle all that apply) A) free weights B) nautilus C) Swim bench  D) surgical tube  25. Do you ever stretch before a workout?  E) other Yes  No  Yes  No  26. At what age did you begin stretching? 27. How many times per week do you stretch? 28. How long do you stretch for each day? 29. What muscles groups do you focus on when stretching?  30. Do you ever use hand paddles while swimming? 31. How often do you use hand paddles?  121  SIS Stroke Analysis Name:  Date:  ID#: Please answer the following questions about each swimmers' Freestyle technique. Circle the most appropriate response. 1. Does this athlete breathe more frequently to one side than the other?  YES If yes, which side?  NO  RIGHT  LEFT  2. How much does this athlete roll his/her body while swimming freestyle?  1 very little or none  2 slightly below average  3 average  4 slightly above average  5 very much or excessive  3. How high does this athlete raise their elbows during the recovery phase of the freestyle stroke?  Right side: 1 very low Left side: 1 very low  2 slightly below average  slightly below average  3 average  3 average  4 slightly above average  5 very high  slightly above average  very high  4. Compared to other swimmers, how many strokes does this athlete require to complete a single length of freestyle?  1 very few  2 slightly less than average  3 average  slightly more than average  very many  122  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0077314/manifest

Comment

Related Items