tendon injuries. basic science and clinical medicine.maffulli, nicola. renstrom, per. leadbetter, wayne
Nicola Maffulli, MD, MS, PhD, FRCS(Orth)Professor and Head, Department of Trauma and Orthopaedic Surgery, KeeleUniversity School of Medicine, Stoke-on-Trent, UKPer Renström, MD, PhDProfessor and Head, Section of Sports Medicine, Department of Surgical Sciences,Karolinska Institute, Stockholm, SwedenWayne B. Leadbetter, MDAdjunct Professor, Uniformed Services University of Health Sciences, F. EdwardHebert School of Medicine, Bethesda, MD, USAEditorsTendon InjuriesBasic Science and Clinical MedicineWith 187 Illustrations, 21 in Full Color 3
PrefaceStandard textbooks of anatomy, physiology, pathology, orthopedic surgery, and sportsmedicine provide little information on tendons. Tendon ailments are increasinglyprevalent in orthopedic surgery and sports medicine, and in occupational and familymedicine as well. This book provides a comprehensive presentation on human tendons for a widerange of readers, from students and teachers of physical education, biomechanics, med-icine, and physical therapy to specialists such as orthopaedic surgeons, pathologists,and physicians specializing in sports medicine. We describe the current principles ofdiagnosis, treatment, and rehabilitation of tendon injuries and disorders. Although weacknowledge that these principles are constantly changing, this book gives readers thetools presently available to the scientiﬁc and biomedical community to tackle tendonproblems. This book has been conceived to be used as a comprehensive source forphysicians, surgeons, physical therapists, chiropractors, sports coaches, athletes, ﬁtnessenthusiasts, and students in a variety of disciplines. The book is deﬁnitely a medical book, but with appeal to professionals outside themedical ﬁeld. The editors have collectively more than 70 years of experience in orthopaedic sportsmedicine, and have dedicated much of their research efforts to studying the patho-physiology of tendon problems. We believe that, as a team, our knowledge and expe-rience will give help and guidance in the management of tendon problems. In recent years—at least in the West—the demand for heavy physical work hasmarkedly decreased. Conversely, leisure-time sports activities have become morepopular, frequent, and intense. Repetitive work, excessive weight, poor ﬁtness, and thelack of regular exercise and of variation in physical loading have all contributed to theincreased incidence of degenerative changes in the musculoskeletal system. Tendonproblems are seen frequently in nonathletes. Modern athletes also suffer from tendonailments. The biological limits that musculoskeletal tissues can withstand are exceeded,with overuse and acute injuries, especially in tendons. This book provides principles of diagnosis, treatment, and rehabilitation for varioustendon problems. We envisage the book to be heavily used by physicians, surgeons,physical therapists, athletic trainers, and other professionals treating patients withtendon problems. We would not have been able to write this book without the help of our coauthorsfrom all over the world. To them, our thanks and appreciation. Nicola Maffulli, MD, MS, PhD, FRCS(Orth) Per Renström, MD, PhD Wayne B. Leadbetter, MD v
ContentsPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vList of Principal Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiPart I Basic Sciences, Etiology, Pathomechanics, and Imaging 1 Anatomy of Tendons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Moira O’Brien 2 Mechanical Properties of Tendons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Constantinos N. Maganaris and Marco V. Narici 3 Growth and Development of Tendons . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Laurence E. Dahners 4 Aging and Degeneration of Tendons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Pekka Kannus, Mika Paavola, and Lászlo Józsa 5 Epidemiology of Tendon Problems in Sport . . . . . . . . . . . . . . . . . . . . . . . 32 Mika Paavola, Pekka Kannus, and Markku Järvinen 6 Neurogenic, Mast Cell, and Gender Variables in Tendon Biology: Potential Role in Chronic Tendinopathy . . . . . . . . . . . . . . . . . . . . . . . . . . 40 David A. Hart, Cyril B. Frank, Alison Kydd, Tyler Ivie, Paul Sciore, and Carol Reno 7 Imaging of Tendon Ailments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Tudor H. HughesPart II Anatomical Sites and Presentation 8 Injury of the Musculotendinous Junction . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Jude C. Sullivan and Thomas M. Best 9 Insertional Tendinopathy in Sports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 Per Renström and Thomas Hach10 Tendon Avulsions in Children and Adolescents . . . . . . . . . . . . . . . . . . . . . 86 Sakari Orava and Urho Kujala vii
viii Contents11 Tendinopathy in the Workplace . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Leo M. Rozmaryn12 Rotator Cuff Tendinopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Andrew Carr and Paul Harvie13 Rotator Cuff Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Theodore A. Blaine and Louis U. Bigliani14 Tendinopathies Around the Elbow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 Alan J. Johnstone and Nicola Maffulli15 Hand and Wrist Tendinopathies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Graham Elder and Edward J. Harvey16 Groin Tendon Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 Per Renström17 Knee and Thigh Overuse Tendinopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Barry P. Boden18 Patellar Tendinopathy and Patellar Tendon Rupture . . . . . . . . . . . . . . . . . 166 Karim M. Khan, Jill L. Cook, and Nicola Maffulli19 Hindfoot Tendinopathies in Athletes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 Francesco Benazzo, Mario Mosconi, and Nicola Maffulli20 Achilles Tendon Rupture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Deiary Kader, Mario Mosconi, Francesco Benazzo, and Nicola Maffulli21 Achilles Tendinopathy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Deiary Kader, Nicola Maffulli, Wayne B. Leadbetter, and Per RenströmPart III Management of Tendon Injuries22 Anti-Inﬂammatory Therapy in Tendinopathy: The Role of Nonsteroidal Drugs and Corticosteroid Injections . . . . . . . . . . . . . . . . . . . 211 Wayne B. Leadbetter23 The Effect of Therapeutic Modalities on Tendinopathy . . . . . . . . . . . . . . . 233 Jason D. Leadbetter24 Rehabilitation After Tendon Injuries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Sandra L. Curwin25 Surgery for Chronic Overuse Tendon Problems in Athletes . . . . . . . . . . . 267 Nicola Maffulli, Per Renström, and Wayne B. LeadbetterPart IV New Developments26 Research Methodology and Animal Modeling in Tendinopathy . . . . . . . . 279 Joanne M. Archambault and Albert J. Banes27 Tendon Innervation and Neuronal Response After Injury . . . . . . . . . . . . 287 Paul W. Ackermann, Daniel K-I. Bring, and Per Renström
Contents ix 28 The Use of Growth Factors in the Management of Tendinopathies . . . . . 298 Louis C. Almekinders and Albert J. Banes 29 Optimization of Tendon Healing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 Nicola Maffulli and Hans D. Moller 30 Gene Therapy in Tendon Ailments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Vladimir Martinek, Johnny Huard, and Freddie H. Fu 31 Tendon Regeneration Using Mesenchymal Stem Cells . . . . . . . . . . . . . . . 313 Stephen Gordon, Mark Pittenger, Kevin McIntosh, Susan Peter, Michael Archambault, and Randell Young Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
List of Principal ContributorsPaul W. Ackermann, MDOrthopedic Laboratory, Research Center, Karolinska Hospital, S-171 76, Stockholm,SwedenLouis C. Almekinders, MDClinical Professor, North Carolina Orthopaedic Clinic, Duke University HealthSystem, Durham, NC 27704, USAAlbert J. Banes, MDDirector of Research, Department of Orthopaedics, University of North Carolina atChapel Hill, Chapel Hill, NC 27599-7052, USAThomas M. Best, MDAssociate Professor of Orthopedics and Rehabilitation and Family Medicine,University of Washington Medical School, Madison, WI 53711, USATheodore A. Blaine, MDAssociate Director, Center for Shoulder, Elbow, and Sports Medicine, Co-Director,Columbia Center for Orthopaedic Research, Columbia University Department ofOrthopaedics, New York, NY 10032, USABarry P. Boden, MDAdjunct Assistant Professor of Surgery, The Uniformed Services University of theHealth Sciences, The Orthopaedic Center, Rockville, MD 20850, USAAndrew Carr, MDNufﬁeld Department of Orthopaedic Surgery, Nufﬁeld Orthopaedic Centre NHSTrust, Headington, Oxford OX3 7LD, UKSandra L. Curwin, MDDepartment of Physical Therapy, University of Alberta, Edmonton, AB, Canada T6G2G4Laurence E. Dahners, MDProfessor of Orthopaedics, University of North Carolina, Chapel Hill, NC 27599, USAStephen Gordon, MDVP, Strategic Planning, Cognate Therapeutics Inc., Bethesda, MD 20814, USA xi
xii List of Principal ContributorsDavid A. Hart, MDMcCaig Centre for Joint Injury and Arthritis Research, Faculty of Medicine,University of Calgary, Calgary, AB, Canada T2N 4N1Edward J. Harvey, MDMcGill University Health Centre, Division of Orthopaedic Surgery, Montreal GeneralSite, Montreal QC, Canada H3G 1A4Tudor H. Hughes, MDAssociate Professor of Radiology, Department of Radiology, University of California,San Diego, Medical Center, San Diego, CA 92013-8756, USAMarkku Järvinen, MDDepartment of Medicine, Tampere University, FIN-33101 Tampere, FinlandPekka Kannus, MDAccident and Trauma Research Center and Tampere Research Center of SportsMedicine, UKK Institute, FIN-33500 Tampere, FinlandJason D. Leadbetter, MDThe Orthopaedic Center, P.A., Rockville, MD 20850, USAWayne B. Leadbetter, MDAdjunct Professor, Uniformed Services University of Health Sciences, F. EdwardHerbert School of Medicine, Bethesda, MD, and The Orthopaedic Center, P.A.,Rockville, MD 20850, USANicola Maffulli, MD, MS, PhD, FRCS(Orth)Professor and Head, Department of Trauma and Orthopaedic Surgery, Keele Univer-sity School of Medicine, North Staffordshire Hospital, Thornburrow Drive, Hartshill,Stoke-on-Trent, Staffordshire, ST4 7QB UKConstantinos N. Maganaris, MDCentre for Biophysical and Clinical Research into Human Movement, ManchesterMetropolitan University, UKVladimir Martinek, MDAssistant Professor: Department of Orthopaedic Sports Medicine, Technical Univer-sity Munich, Munich, GermanyMoira O’Brien, MDProfessor, Human Performance Laboratory, Department of Anatomy, Trinity College,Dublin 2, IrelandSakari Orava, MD, PhDProfessor, Mehilainen Hospital and Sports Clinic, 20100 Turku, FinlandPer Renström, MD, PhDProfessor and Head, Section of Sports Medicine, Department of Surgical Sciences,Karolinska Hospital, SE 171 76 Stockholm, SwedenLeo M. Rozmaryn, MDThe Orthopaedic Center, P.A., Rockville, MD 20850, USA
Part I Basic Sciences, Etiology,Pathomechanics, and Imaging
1Anatomy of TendonsMoira O’BrienA tendon forms an integral part of a musculotendinous sipated laterally, relative to the axis of the tendon. Theunit. Its primary function is to transmit forces from occupation and sports activity of the individual may altermuscle to rigid bone levers producing joint motion [1,2]. the alignment of the ﬁbers of the tendon.Tendons are stronger than muscles, are subjected to both The majority of the ﬁbers run in the direction of stresstensile and high compressive forces, and can sustain 17  with a spiral component, and some ﬁbers run perpen-times body weight. They act as shock absorbers, energy dicular to the line of stress . Small-diameter ﬁbers maystorage sites, and help to maintain posture through their run the full length of a long tendon , but ﬁbers with aproprioceptive properties . High rates of loading make diameter greater than 1500 Å may not extend the fulltendons more brittle, thus absorbing less energy, but length of a long tendon .being more effective moving heavy loads . The con- The details of the gross anatomy of some tendons haveverse occurs at low rates of loading, when tendons are been known for some time, but the ﬁner details and vari-more viscous, absorb more energy, and are less effective ations of a large number of tendons have not often beenat moving loads . emphasized. For example, the spiral arrangement of the Tendons generally tend to concentrate the pull of a ﬁbers of the tendon of ﬂexor digitorum superﬁcialis asmuscle on a small area. This enables the muscle to change they ﬂatten, fork, and fold around the ﬂexor digitorumthe direction of pull and to act from a distance. A tendon profundus to allow it to reach its insertion into the distalalso enables the muscle belly to be at an optimal distance phalanx of the hand and the similar arrangement of thefrom a joint without requiring an extended length of ﬂexor digitorum brevis and the longus in the foot havemuscle between the origin and insertion. only recently been clariﬁed (see Figure 1-3). The range of motion of a musculotendinous unit and Tendons were usually described as having a parallelthe force applied to the tendon determine the orientation orientation of collagen ﬁbers  until transmission andof the ﬁbers, relative to the axis of the tendon.The greater scanning electron microscopy demonstrated that colla-the longitudinal array of the muscle ﬁbers, the greater the gen ﬁbrils are orientated longitudinally, transversely, andrange of motion of the muscle and the tendon. The horizontally. The longitudinal ﬁbrils cross each other,strength of a tendon depends on the number, size and forming spirals and plaits [11,12]. Transmission andorientation of the collagen ﬁbers. It also depends on the scanning electron microscopy have demonstrated thatthickness and internal ﬁbrillar organization  (see the interior of the tendon consists mainly of longitudinalFigures 1-1 and 1-2). ﬁbrils with some transverse and horizontal collagen ﬁbrils Collagen ﬁbers are distributed in different patterns. In .tendons, where tension is exerted in all directions, the Tendons vary in shape and size. They may be ﬂattenedﬁber bundles are interwoven without regular orientation, or rounded. They may be found at the origin or insertionand the tissues are irregularly arranged. If tension is in of a muscle, or form tendinous intersections within aonly one direction, the ﬁbers have an orderly parallel muscle. An aponeurosis is a ﬂattened tendon, consistingarrangement, i.e. are regularly arranged. In most regions, of several layers of densely arranged collagen ﬁbers. Thecollagenous ﬁbers are the main component. fascicles are parallel in one layer but run in different Fusiform muscles exert greater tensile force on their directions in adjacent layers. The aponeurosis may formtendons than pennate muscles because all the force is a major portion of a muscle, e.g. the external oblique,applied in series with the longitudinal axis of the tendon. internal oblique, and transversus abdominis muscles. TheThe more oblique the muscle ﬁbers, the more force is dis- aponeurosis of the external oblique forms part of the 3
4 M. O’BrienFigure 1-1. (A) Diagram of the inferior attachment of a tendonshowing plaited component ﬁbers. (B and C) Different ﬁberstake the strain in different positions of a joint.rectus sheath, the inguinal ligament, and lacunar liga-ments. The aponeurosis of the internal oblique and trans-versus form the conjoint tendon, which takes part in theformation of the lower portion of the anterior wall of therectus sheath and the medial part of the posterior wall ofthe inguinal canal. The bicipital aponeurosis of the bicepsbrachii extends its insertion into the ulna. Laminatedtendons are found in the pectoralis major, latissimusdorsi, and masseter muscles. Tendons may give rise to ﬂeshy muscles, e.g. the lum- Figure 1-2. Multipennate.bricals, arising from the ﬂexor digitorum profundustendons in the hand and the ﬂexor digitorum longus in cartilaginous nodules in the fetus. In the upper limb,the foot. The oblique ﬁbers of the vastus medialis arise sesamoid bones are found on the palmar aspect in thefrom the tendon of the adductor magnus. The oblique upper limb, in the insertion of the two heads of the adduc-ﬁbers of the vastus lateralis arise from the iliotibial tract. tor pollicis on the ulnar side, and in the ﬂexor pollicisThe semimembranosus tendon has several expansions brevis at its insertion into the radial side of the base ofthat form ligaments including the oblique popliteal liga- the proximal phalanx of the thumb. The pisiform is ament of the knee and the fascia covering the popliteus sesamoid in the tendon of the ﬂexor carpi ulnaris. Amuscle (Figure 1-4). sesamoid is occasionally found in the biceps brachii Segmental muscles that develop from myotomes often tendon in relation to the radial tuberosity.have tendinous intersections. In certain areas each The patella in the tendon of the quadriceps is thesegment has its own blood and nerve supply. These largest sesamoid in the body (see Figure 1-5). There isinclude the rectus abdominis, the hamstrings, and the occasionally a sesamoid in the lateral head of the gas-sternocleidomastoid. trocnemius (fabella), in the tibialis anterior, opposite the Sesamoid bones may develop in tendons where they distal aspect of the medial cuneiform, or in the tibialiscross articular surfaces or bone: They are present as posterior below the plantar calcaneonavicular ligament,Figure 1-3. Flexor digitorum superﬁcialis ﬂattens, forks, and folds to allow ﬂexor digitorum profundus to insert into distal phalanx.
1. Anatomy of Tendons 5Figure 1-4. Lumbricals arising from tendons of ﬂexor digito-rum profundus in the hand.the spring ligament . A sesamoid may occur in theperoneus longus tendon before it enters the groove in thecuboid. There are always two sesamoid bones associatedwith the insertion of the ﬂexor hallucis brevis.The medial,the larger, is found in the abductor hallucis and themedial half of the ﬂexor hallucis brevis. The lateral is inthe combined insertion of the lateral half of the ﬂexorhallucis brevis and the adductor hallucis. The medialsesamoid may be bipartite, usually a bilateral feature (see Figure 1-5). Tendons may be intracapsular, e.g. the long head of thebiceps brachii and the popliteus. The synovial membraneof the joint surrounds the tendons inside the joint andextends for a variable distance beyond the joint itself. The knowledge of the extent of the synovial cover- Figure 1-5. Patella in quadriceps tendon.ing is important when deciding to inject around a joint.The synovial sheath, which surrounds the long head ofthe biceps brachii, extends to the lower border of the of two continuous, concentric layers, which are separatedlatissimus dorsi insertion, approximately the lower by a ﬁlm of ﬂuid. The visceral layer surrounds the tendon,border of the posterior fold of the axilla. and the parietal is attached to the adjacent connective Tendons are covered by ﬁbrous sheaths, or retinacula, tissues. As a tendon invaginates into the sheath, there isas they pass over bony prominences or lie in grooves often a mesotendon.lined with ﬁbrocartilage to prevent them from bow-stringing when the muscle contracts . Reﬂectionpulleys hold tendons as they pass over a curved area, e.g.the transverse humeral ligament that holds the long headof the biceps as it leaves the shoulder joint and the supe-rior and inferior peroneal retinacula surrounding the per-oneus longus and peroneus brevis. Fibrocartilage waspresent in 22 of 38 tendon sites where tendons pressedagainst bone . Most retinacula are mainly ﬁbrous, butthe inferior peroneal retinaculum and the trochlear reti-naculum in the orbit for the superior oblique muscle arecartilaginous  (see Figure 1-6). When tendons run in ﬁbro-osseous tunnels or passunder retinacula, fascial slings bind them down; they areenclosed in synovial membrane. The membrane consists Figure 1-6. Extensor retinaculum of wrist.
6 M. O’Brien Synovial folds in the ﬁbro-osseous sheaths of the pha- tendinous ﬁbers is tailored to direct the force generatedlanges of the hand and foot are called the vincula longa by the muscular contraction to the point of insertion.and vincula brevia. They contain the blood vessels that The musculotendinous junction is considered thesupply the ﬂexor tendons inside the sheaths. The longa growth plate of muscle, as it contains cells that can elon-are thinner, and are found proximally; the brevia are gate rapidly and deposit collagen. The tendon elongatesshorter, and are found at the insertions of the tendons. here. It is a complex area that contains the organs ofThe lining of the sheath is extremely cellular and vascu- Golgi and nerve receptors. The muscle ﬁbers may showlar. It secretes synovial ﬂuid, and reacts to inﬂammation terminal expansions. Electron microscopy shows thatby cellular proliferation and the formation of more ﬂuid. these ends have a highly indented sarcolemma, with aThis may result in adhesions and restriction of movement dense internal layer of cytoplasm into which the actin ﬁl-between the two layers. aments of the adjacent sarcomeres are inserted . The Bursae are associated with many tendons and help to basement membrane is prominent, and the collagen andreduce friction between 1) tendons, e.g. the tibial inter- reticulum ﬁbers lie in close contact. Subsarcolemmaltendinous bursae at the insertions of the tendons of deposits of dystrophin occur at the junctional folds andsartorius, gracilis, and semitendinosus; 2) tendons and the extrajunctional sarcolemma of the myotendinousaponeurosis, e.g. the gluteus maximus and aponeurosis of junction, suggesting that dystrophin may be one of thevastus lateralis; 3) tendons and bone; 4) deep infrapatel- compounds linking terminal actin ﬁlaments to the sub-lar bursae, e.g. the ligamentum patellae and tibial plasmalemmal surface of the junctional folds of thetuberosity, subacromial bursa, and retrocalcaneal bursa. myotendon .The olecranon bursa and the superﬁcial infrapatellar Muscle tears tend to occur at the musculotendinousbursa are examples of bursae between tendons and skin. attachments . Variations in the extent of the tendon Arthroscopy, magnetic resonance imaging (MRI), and into the muscle at the origin and insertion may explainultrasound have emphasized the prevalence of variations the site of muscle tears. There are variations in the shapein muscles and tendons. The variations in the anatomy and extent of the adductor longus tendon. Tendinousmay affect the entry of an arthroscope or cause difﬁculty intersections are found in the hamstrings denoting thein interpretation of MRI studies. The attachments of the original myotomes  (see Figure 1-7).long head of the biceps to the supraglenoid tubercle andthe superior margin of the glenoid labrum are intracap-sular, and may be involved in a Type IV superior labrumanterior-posterior (SLAP) lesion, when there is a bucket-handle tear of the superior labrum with extension of thetear into the biceps tendon . Supernumerary tendons may occur. The most commontendon in the lower limb to have an accessory tendon isthe soleus muscle-tendon complex. When present, it mayhave its own tendon of insertion anterior to the soleus. The plantaris may also be duplicated. Supernumerarytendons have been reported in the tibialis anterior, tib-ialis posterior and peroneus longus . The plantaris inthe leg and the palmaris longus in the forearm are themost frequent tendons that may be absent.Musculotendinous JunctionTendons develop independently in the mesenchyme, andtheir connection with their muscle is secondary. Themyotendinous junction is the junctional area between themuscle and the tendon and is subjected to great mechan-ical stress during the transmission of muscular contractileforce to the tendon . The extension of a tendon’s col-lagen ﬁbers into the body of the muscle increases theanchoring surface area . It can continue as a single oras multiple visible structures or as a diffuse network,visible only under a microscope. The arrangement of the Figure 1-7. Musculotendinous junction of adductor longus.
1. Anatomy of Tendons 7Osteotendinous Junction the supraspinatus. The ﬁbrocartilage acts as a stretching brake, as a stretched tendon tends to narrow, but the car-The insertion of a tendon into bone, or the osteotendi- tilage matrix prevents this so that it does not stretch atnous junction (OTJ), involves a gradual transition from its interface with bone. The structure of the attachmenttendon to ﬁbrocartilage to lamellar bone, and consists of zone of a tendon may vary, depending on the occupation4 zones of pure ﬁbrous tissue, unmineralized ﬁbrocarti- and sports activity of the individual . The insertion oflage, mineralized ﬁbrocartilage, and bone . There are the biceps of a window cleaner, who works with hisone or more prominent basophilic lines (cement or blue forearm pronated, would differ from that of an individ-lines), called the tidemark. The tidemark represents the ual who works with the forearm supinated.outer limit of the mineralized ﬁbrocartilage. The line isusually smoother than at the osteochondral junction.Chondrocytes are found on the tendon side of the tide- Nerve Supplymark, and tendon ﬁbers can extend as far as the osteo-chondral junction. Very few blood vessels cross from Tendons are supplied by sensory nerves from the overly-bone to tendon. Collagen ﬁbers often meet the tidemark ing superﬁcial nerves or from nearby deep nerves. Theat right angles, i.e. there is a change in the angle just nerve supply is largely, if not exclusively, afferent. Thebefore the tendon becomes cartilaginous, and only a afferent receptors are found near the musculotendinousgradual change occurs inside the ﬁbrocartilage. If the junction , either on the surface or in the tendon. Theattachment is very close to the articular cartilage, the nerves tend to form a longitudinal plexus and enter viazone of ﬁbrocartilage is continuous with the articular car- the septa of the endotenon or the mesotendon if there istilage. Under electron microscopy, it is found to be com- a synovial sheath. Branches also pass from the paratenonposed of densely packed, randomly oriented collagen via the epitenon to reach the surface or the interior of aﬁbrils of varying diameters that are continuous with those tendon .of the unmineralized and mineralized ﬁbrocartilage. The There are 4 types of receptors. Type I receptors, calledchemical composition of ﬁbrocartilage is age dependent, Rufﬁni corpuscles, are pressure receptors that are veryboth in the OTJ and other ﬁbrocartilaginous zones of the sensitive to stretch and adapt slowly . Rufﬁni corpus-tendon. cles are oval and 200 mm by 400 mm in diameter. Type II Osteogenesis at a tendon-bone junction allows a receptors, the Vater-Pacini corpuscles, are activated bysmooth mechanical transition. Periosteum is specialized, any movement. Type III receptors, the Golgi tendondense connective tissue, and has an outer vascularized organs, are mechanoreceptors. They consist of unmyeli-layer that is mostly ﬁbrous, and an inner cellular layer. It nated nerve endings encapsulated by endoneural tissue.possesses osteogenic potential, except where tendons are They lie in series with the extrafusal ﬁbers and monitorinserted. The periosteum is connected to the underlying increases in muscle tension rather than length. The Golgibone by dense collagen ﬁbers, extending its outer ﬁbrous tendon organ is 100 mm in diameter and 500 mm in length.layer into the mineralized bone matrix perpendicular to The tendon ﬁber is less compact here than in the rest ofthe bone surface. During bone growth, collagen ﬁbers the tendon. The endoneural tissue encapsulates thefrom the tendon are anchored deeper into the deposited unmyelinated nerve ﬁbers. The lamellated corpusclesbone. Variations in the attachments of tendon to bone respond to stimuli transmitted by the surrounding tissues,may explain the variations in hot spots on bone scans e.g. pressure, which is produced by muscle contraction.when stress fractures are present in the tibia . The amount of pressure depends on the force of con- A tendon can be attached to bone in several ways. The traction. They may provide a more ﬁnely tuned feedback.insertion may be to the epiphysis or to the diaphysis. It Type IV receptors are the free nerve endings that act asmay be a ﬂeshy attachment to the periosteum or a tendi- pain receptors.nous attachment to a bony crest, ridge, or prominence.Fleshy attachments produce smooth, featureless surfacesindistinguishable from areas of bone covered by perios- Blood Supplyteum alone, but attachments of tendons, aponeurosis, andﬁbrous septa produce distinct markings e.g. tubercles or The blood supply of tendons is very variable, and isridges . usually divided into three regions: 1) The musculotendi- There is no periosteum if ﬁbrocartilage is present at the nous junction; 2) the length of the tendon; and 3) thetendon attachment . Benjamin et al.  found that tendon-bone junction. The blood vessels originate frommost tendons attached to the ends of long bones had vessels in the perimysium, periosteum, and via theﬁbrocartilage at their attachments, but the amount of paratenon and mesotendon.ﬁbrocartilage varied. Fibrocartilage was usually most The blood supply to the musculotendinous junction isobvious in the portion of the tendon nearest a joint, e.g. from the superﬁcial vessels in the surrounding tissues.
8 M. O’BrienSmall arteries branch and supply both muscles and Blood vessel within septatendons, but they are completely separate as there is no enclosing tertiary bundlesanastomosis between the capillaries. The main blood supply to the middle portion of thetendon is via the paratenon. In tendons that are exposedto friction and are enclosed in a synovial sheath, it is viathe vincula (see Figure 1-8). The small blood vessels inthe paratenon run transversely towards the tendon, andbranch several times before running parallel to the longaxis of the tendon. The vessels enter the tendon along theendotenon; the arterioles run longitudinally ﬂanked bytwo venules. Capillaries loop from the arterioles to thevenules, but they do not penetrate the collagen bundles Secondary bundle(see Figure 1-9). Primary bundle Vessels supplying the bone-tendon junction supply the Spaces occuped by tendon cellslower one-third of the tendon. There is no direct com-munication between the vessels because of the ﬁbrocar- Figure 1-9. Transverse section of tendon.tilaginous layer between the tendon and bone, but thereis some indirect anastomosis between the vessels.Tendons that go around corners are subject to greaterstrain, and are more likely to have interference with their tendinous portions and between the anterior and poste-blood supply, particularly if they cross an articular rior vessels. There is now evidence that there is an areasurface, as they may also be subjected to compressive of hypervascularity secondary to low-grade inﬂammationforces, which may result in cartilaginous changes in the with neovascularization due to mechanical irritation intendon from Type I to Type II collagen. the critical zone of the supraspinatus . The blood supply of tendons is compromised at sites of The blood supply of the ﬂexor tendons of the hand canfriction, torsion, or compression. This is found particu- be divided into two regions. The blood supply of the syn-larly in the tibialis posterior, supraspinatus, and Achilles ovial-covered tendons consists of longitudinal vasculartendons [25–27]. There is a characteristic vascular pattern bundles with short transverse anastomosis, while non-in the rotator cuff tendons, with a constant area of reac- synovial-covered tendons with paratenon have a uniformtive avascularity approximately 0.7 to 1 cm from the blood supply. The synovial-covered portions of the ﬂexorinsertion. This critical area is the junction between the digitorum superﬁcialis and the ﬂexor digitorum profun-two groups of blood vessels, supplying the muscular and dus receive their blood supply only on the dorsal aspect. There are avascular regions at the metacarpophalangeal joint and at the proximal interphalangeal joint, possibly resulting from the mechanical forces exerted at these Tendon zones . The long ﬂexor tendons are supplied by two main sources: primarily by small arteries that run in the Sheath vincula longa and brevia and reach the dorsal surface of the tendon; and secondarily by small intrinsic longitudi- Muscle nal vessels that run parallel to the collagen ﬁbers of the tendon and extend from the muscular attachments of the long ﬂexor tendons. The Achilles tendon is supplied at its musculotendi- nous junction, along the length of the tendon, and at its Fluid junction with bone. The blood supply consists mainly of longitudinal arteries that course the length of the tendon. The area of lowest vascularity is 2 to 6 cm above the inser- tion of the tendon. The Achilles tendon is the thickest and the strongest tendon. It is approximately 15 cm long, and Mesotendineum on its anterior surface it receives the muscular ﬁbers from with blood vessels the soleus almost to its insertion. The tendon is at ﬁrst ﬂattened at its junction with the gastrocnemius, and thenFigure 1-8. Blood supply of tendon surrounded by a synovial it becomes rounded. It expands at its insertion, where itsheath. becomes cartilaginous . The soleus and the gastrocne-
1. Anatomy of Tendons 9mius vary in their contribution to the Achilles tendon and are sparse cells. Cross-section of tendons shows inactivein the extent of their fusion. The soleus varies from 3 to ﬁbroblast cells .11 cm, and the gastrocnemius from 11 to 16 cm. As the Five tropocollagen units unite to form ﬁbrils. Severaltendon descends it twists, and the gastrocnemius is found parallel ﬁbrils embedded in the extracellular matrix con-mainly on the lateral and posterior part of the tendon. stitute a ﬁber. A group of ﬁbers constitute a fascicle, theRotation begins above the region where the soleus tends smallest collagenous structure that can be tested .to join, and the degree of rotation is greater if there is Fascicles are surrounded by endotenon, epitenon, andminimal fusion .The twisting produces an area of stress paratenon. The endotenon is a mesh of loose connectivein the tendon, which is most marked 2 to 5 cm above the tissue, which surrounds collagen bundles. The endotenoninsertion, which is the area of poor vascularity and a holds the bundles together, permits some movement ofcommon site of tendon ailments [28–30]. the bundles relative to each other, and carries blood vessels, lymphatics, and nerves. A ﬁne connective tissue sheath, the epitenon, is continuous throughout the innerStructure of Tendons surface with the endotenon, and surrounds the whole tendon . The paratenon is the outermost layer and isTendons appear white, as they are relatively avascular. A composed of loose, fatty, areolar tissue surrounding thetendon is a roughly uniaxial composite, composed mainly tendon: Nerves and blood vessels run through it. Fluidof Type I collagen in an extracellular matrix composed may be found between the paratenon and the epitenon,mainly of mucopolysaccharides and a proteoglycan gel preventing friction . Its mechanical function is to. Tendons consist of 30% collagen and 2% elastin allow the tendon to glide freely against the surroundingembedded in an extracellular matrix containing 68% tissue. The connective tissue that surrounds the ﬁbrils, thewater and tenocytes . Elastin contributes to the ﬂex- fascicles, and the entire muscle consists mainly of Type Iibility of the tendon. The collagen protein tropocollagen collagen, with a minor component consisting of Type IIIforms 65% to 80% of the mass of dry weight tendons and collagen.Type IV collagen is found in the basement mem-ligament (see Figure 1-10). brane, with traces of Type V collagen. Ligaments and tendons differ from other connectivetissues in that they consist mainly of Type I collagen. Lig- Collagen Formationaments have 9% to 12% of Type III collagen, and aremore cellular than tendons . Type II collagen is found The structural unit of collagen is tropocollagen, a long,abundantly in the ﬁbrocartilage at the attachment zone thin protein 280 nm long and 1.5 nm wide, which consistsof the tendon (OTJ) and is also present in tendons that mainly of Type I collagen  (see Figure 1-11).Tropocol-wrap around bony pulleys. Collagen consists of clearly lagen is formed in the ﬁbroblast cell as procollagen, whichdeﬁned, parallel, and wavy bundles. Collagen has a char- is then secreted and cleaved extracellularly to becomeacteristic reﬂective appearance under polarized light. collagen. The 100 amino acids join to form an alpha-Between the collagen bundles, fairly evenly spaced there chain. There are 3 alpha-chains, which are surrounded by a thin layer of proteoglycans and glycosaminoglycans. Two of the alpha-chains are identical (alpha-1), and one differs slightly (alpha-2). The three-polypeptide chains each form a left-handed helix. The chains are connected by hydrogen bonds and wind together to form a ropelike, right-handed superhelix , which gives the collagen molecule a rodlike shape . Almost two-thirds of the collagen molecule consists of 3 amino acids: glycine (33%), proline (15%), and hydroxyproline (15%). Each alpha-chain consists of a repeating triplet of glycine and two other amino acids. Glycine is found at every third residue, while proline (15%) and hydroxyproline (15%) occur frequently at the other two positions. Glycine enhances the stability by forming hydrogen bonds among the 3 chains. Collagen also contains two amino acids, hydroxyproline and hydroxylysine (1.3%), not often found in other proteins . The ﬁrst stage in the synthesis of collagen is the for- mation inside the cell of mRNA for each type of the Figure 1-10. Schematic drawing of a tendon. polypeptide alpha-chain. The polypeptide alpha-chains
10 M. O’Brien D collagen there are nonhelical peptides, the domains.A When procollagen leaves the cell, the domains areFibril cleaved enzymatically by peptides to form tropocollagen. The adjacent molecules of collagen pack together over- Overlap Zone 0.4 D Microfibrils lapping by a quarter stagger, and appear as cross-stria- tions under an electron microscope . Hole Zone 0.6 DBPacking of CrosslinksMolecules Tropocollagen molecules are stabilized and held together by electrostatic, crosslinking chemical bonds. Hydrox- yproline is involved in hydrogen bonding (intramolecu- larly) between the polypeptide chains. Hydroxylysine is DC 3 000 A (4.4 D) involved in covalent (intermolecularly) crosslinkingCollagenMolecule between adjacent tropocollagen molecules . Both 15 A Diameter increase the strength of collagen, and the crosslinks result 104 A (0.15 D) a2DTriple a1 IHelix N a1 T OH R OH A C E L L OH OH 12.4 A UE LTypical Clycine Hydroxyproline A Procollagen MoleculeSequence in Ra1 and a2 HOChains x OH OH Proline OH OH Figure 1-11. Tropocollagen. Tropocollagen Molecule E X T assembly into R microfibrilassemble on the polyribosomes that are bound to the A Cmembranes of the rough endoplasmic reticulum. They Eare then injected into the cisternae as preprocollagen L Lmolecules. The signal peptide is clipped off, forming pro- Ucollagen. About half the proline and some lysine are L Ahydroxylated inside the tenoblast, just before the chains R Microfibriltwist into the triple helix to form procollagen. Theenzymes that mediate this require iron and vitamin C ascofactors. Cross-linking Hydroxyproline is involved in the hydrogen bondingbetween the polypeptide chains, while hydroxylysine isinvolved in the covalent crosslinking of tropocollageninto bundles of various sizes. Both these amino acidsincrease the strength of collagen. In vitamin C deﬁciency, Collagenthere is an excessive amount of hydroxyproline in theurine, and the collagen is defective. At both ends of pro- Figure 1-12. Production of Collagen.
1. Anatomy of Tendons 11from enzyme-mediated reactions, mainly lysine and Ground Substancehydrolysine. The key enzyme is lysyl-oxidase, which is therate-limiting step for collagen crosslinking. Ground substance is a complex mixture of proteoglycans Hydroxylysins containing crosslinks are the most and glycoproteins surrounding the collagen ﬁbers. Itprevalent intermolecular crosslinks in native insoluble has a high viscosity that provides the structural support,collagen. Crosslinks are important to the tensile strength lubrication, and spacing of the ﬁbers essential for glidingof collagen, allow increased energy absorption, and in- and cross-tissue interactions. The ground substance iscrease its resistance to proteases. a medium for the diffusion of nutriments and gases, Collagen ﬁbers acquire all the crosslinks they will have and regulates the extracellular assembly of procollagenshortly after synthesis. Crosslinks are at the maximum in into mature collagen. Water makes up 60% to 80% of theearly postnatal life and reach their minimum at physical total weight of the ground substance. Proteoglycans andmaturity. Newly synthesised collagen molecules are glycoproteins in the ground substance account for lessstabilized by reducible crosslinks, but their numbers than 1% of the total dry weight of tendon. They maintaindecrease during maturation. Nonreducible crosslinks are the water within the tissues and are involved withfound in mature collagen, which is a stiffer, stronger, and intermolecular and cellular interactions. Proteoglycansmore stable. Reduction of crosslinks results in extremely and glycoproteins also play an important role in the for-weak, friable collagen ﬁber. Crosslinking of collagen is mation of ﬁbrils and ﬁbers. The covalent crosslinksone of the best biomarkers of aging. between the tropocollagen molecules reinforce the ﬁbril- Crosslinking substances are produced as charged lar structure.groups, and they are removed by metabolic processes in The water-binding capacity of these macromolecules isearly life but accumulate in old age, e.g. hydroxyproline important. Most proteoglycans are oriented at 90 degreesis released quickly and in large quantities in young to collagen, and each molecule of proteoglycans cananimals, but it is released more slowly and in smaller interact with 4 collagen molecules. Others are randomlyamounts in older animals. arranged to lie parallel to the ﬁbers, but they interact only with that ﬁber . The matrix is constantly being turned over and remodeled by the ﬁbroblasts and by degradingElastin enzymes (collagenases, proteoglycanase, glycosaminogly-Elastin contributes to the ﬂexibility of a tendon. This canase, and other proteases).protein does not contain much hydroxyproline or lysine, The proteogylcans and glycoproteins consist of twobut is rich in glycine and proline. It has a large content of components, glycosaminoglycans (GAGs) and structuralvaline and contains desmosine and isodesmonine, which glycoproteins. The main proteogylcans in tendons associ-form crosslinks between the polypeptides, but no hydrox- ated with glycosaminoglycans are dermatan sulfate,ylysine. Elastin does not form helices and is hydrophobic. hyaluronic sulfates, chondroitin 4 sulfates, and chon-Elastin is usually less than 1 mm in length, has no period- droitin 6 sulfates. Other proteoglycans found in tendonsicity and requires special staining. Very little elastin is include biglycan, decorin, and aggrecan. Aggrecan is afound in healing wounds. chondroitin sulfate bearing large proteoglycan in the ten- sional regions of tendons . The glycoproteins consist mainly of proteins, such as ﬁbronectin, to which carbo-Cells hydrates are attached.The cell types in tendons are tenocytes and tenoblasts or Fibronectins are high-molecular-weight, noncollage-ﬁbroblasts. Tenocytes are ﬂat, tapered cells, spindle- nous extracellular glycocoproteins. Fibronectin plays ashaped longitudinally and stellate in cross section. Teno- role in cellular adhesion (cell-to-cell and cell-to-cytes lie sparingly in rows between collagen ﬁbrils . substrate) and in cell migration. Fibronectin may beThey have elaborate cell processes that form a three- essential for the organization of collagen I and III ﬁbrilsdimensional network extending through the extracellular into bundles, and may act as a template for collagen ﬁbermatrix. They communicate via cell processes and may be formation during the remodeling phase.motile [40,41]. Tenoblasts are spindle-shaped or stellate Hyaluronate is a high-molecular-weight matrix gly-cells with long, tapering, eosinophilic ﬂat nuclei. cosaminoglycan, which interacts with ﬁbronectin toTenoblasts are motile and highly proliferative. They have create a scaffold for cell migration. It later replaceswell-developed, rough endoplasmic reticulum, on which ﬁbronectin.the precursor polypeptides of collagen, elastin, proteo- Integrins are extracellular matrix binding proteins withglycans, and glycoproteins are synthesized . Tendon speciﬁc cell surface receptors. Large amounts of aggrecanﬁbroblasts (tenoblasts) in the same tendon may have dif- and biglycan develop at points where tendons wrapferent functions. The epitenocyte functions as a modiﬁed around bone and are subjected to compressive and ten-ﬁbroblast with well-developed capacity of repair. sional loads. TGF-beta could be involved in differentia-
12 M. O’Briention of regions of tendon subjected to compression, Referencesbecause compressed tendon contains both decorin andbiglycan, whereas tensional tendons contain primarily 1. Robert L, Moczar M, Robert M. (1974) Biogenesis, matu-decorin . ration and aging of elastic tissue (abstract). Experientia. 30:211–212. The synthesis of proteoglycans begins in the rough 2. Kvist M. (1991) Achilles tendon injuries in athletes. Sportsendoplasmic reticulum, where the protein portion is syn- Med. 18(3):173–201.thesized. Glycosylation starts in the rough endoplasmic 3. Benjamin M, Qin S, Ralphs JR. (Dec. 1995) Fibrocartilagereticulum and is completed in the Golgi complex, where associated with human tendons and their pulleys. J Anat.sulfation takes place. The turnover of proteoglycans is 187(Pt):625–633.rapid, from 2 to 10 days. Lysosomal enzymes degrade the 4. Fyfe I, Stanish WD. (1992) The use of eccentric training andproteoglycans, and lack of speciﬁc hydrolases in the stretching in the treatment and prevention of tendonlysososmes results in their accumulation. injuries. Clin Sports Med. 11(3):601–624. When newly formed, the ground matrix appears vac- 5. Oxlund CE. (1986) Relationships between the biomechan-uolated. The formation of tropocollagen and extracellu- ical properties, composition and molecular structure of con-lar matrix are closely interrelated. The proteoglycans in nective tissues. Conn Tiss Res. 15:65–72. 6. Frost HM. (1990) Skeletal structural adaptations tothe ground substance seem to regulate ﬁbril formation as mechanical usage (SATMU), 4: Mechanical inﬂuences onthe content of proteoglycans decreases in tendons when intact ﬁbrous tissue. Anat Rec. 226:433–439.the tropocollagen has reached its ultimate size. An ade- 7. Jozsa L, Kannus P, Balint JB, Reffy A. (1991) Three-dimen-quate amount of ground substance is necessary for the sional structure of tendons. Acta Anat. 142:306–312.aggregation of collagenous proteins into the shape of 8. Kirkendall DT, Garrett WE. (1997) Function and biome-ﬁbrils. chanics of tendons. Scand J Med Sci Sports. 7:62–66. 9. Jozsa L, Kannus P. (1997) Human Tendons: Anatomy, Phys- iology, and Pathology. Champaign, IL: Human Kinetics. 10. Arai H. (1907) Die Blutgefasse der Sehnen. Anat Hefte.Crimp 34:363–382.Collagen ﬁbrils in the rested, nonstrained state are not 11. Chansky HA, Iannotti I P. (1991) The vascularity of the rotator cuff. Clin Sports Med. 10:807–822.straight but wavy or crimped. Crimp represents a regular 12. Jozsa L, Kannus P, Balint BJ, Reffy A. (1991) Three-dimen-sinusoidal pattern in the matrix. Crimp is a feature of sional Ultra structure of human tendons. Acta Anat.both tendons and ligaments. The periodicity and ampli- 142:306–312.tude of crimp is structure speciﬁc . It is best evaluated 13. Williams PC, Warwick R, Dyson M, Bannister L, eds. (1993)under polarized light. Crimp provides a buffer in which Gray’s Anatomy. 37th Ed. London: 651.slight longitudinal elongation can occur without ﬁbrous 14. Warwick R, Williams PC, eds. (1973) Gray’s Anatomy. 35thdamage, and acts as a shock absorber along the length of Ed. Edinburgh, Scotland: Longmans Green and Company;the tissue. Different patterns of crimping exist: straight, 231–232.or undulated in a planar wave pattern. 15. Ippolito E, Postacchini F. (1986) Anatomy. In: Perugia Collagen production can be affected by many factors. L, Postacchini F, Ippolito E, eds. The Tendons: Biology-These include: heredity, diet, nerve supply, inborn errors, Pathology-Clinical Aspects. Milan, Italy: Editrice Kurtis; 9–36.and hormones. Corticosteroids are catabolic, and they 16. Ruland LJ, Matthews LS. (1995) Gross arthroscopicalso inhibit the production of new collagen. Insulin, estro- anatomy in athletic injuries of the shoulder. Editor Pettr-gen and testosterone can actually increase the production noe FA, New York: McGraw-Hill; 1–17.of collagen. 17. Gardner DC, Dodds DC (1976) Human Histology. Edin- Disorders of collagen include osteogenesis imperfecta, burgh, Scotland: Churchill Livingstone; 364–377.Ehlers-Danlos, scurvy, and progressive systemic sclerosis. 18. Garrett WE. (1990) Muscle strain injuries: clinical and basicMuscles and tendons atrophy and the collagen content aspects. Med Sci Sports Exerc. 22:436–443.decreases when the nerve supply to the tendon is 19. Lee C, O’Brien M. (Mar. 1988) Site of the tendinousinterrupted. Inactivity also results in increased collagen interruption in semitendinosus in man. J Anat. 157:229–degradation, decreased tensile strength, and decreased 231.concentration of metabolic enzymes. Due to the reduc- 20. Benjamin M, Evans EJ, Cope L. (1986) The histology of tendon attachment to bone in man. J Anat. 149:89–100.tion of enzymes that are essential for the formation 21. Ekenman I, Tsai-Fellander L, Johansson C, O’Brien M.of collagen with age, repair of soft tissue is delayed in (1995) The plantar ﬂexor muscle attachments on the tibia.the older age groups. Exercise increases collagen Scand J Med Sci Sports. 5:160–164.synthesis, the number and size of the ﬁbrils, and the con- 22. Schneider H. (1959) Die Abnutzungerkrankungen der sehnecentration of metabolic enzymes. Physical training unde ihr Therapie. Stuttgart, Germany: G. Thieme.increases the tensile and maximum static strength of 23. Stilwell DL Jr. (1957) The innervation of tendons andtendons. aponeurosis. Am J Anat. 100:289.
1. Anatomy of Tendons 1324. Freeman MAR,Wyke B. (1967) The innervation of the knee 36. Teitz CC. (1989) Overuse injuries. In: Teitz CC, ed. Scientiﬁc joint: an anatomical and histological study in the cat. J Anat. Foundations of Sports Medicine. Toronto: B.C. Decker; 101:505–532. 299–325.25. Frey C, Shereff M, Greenidge N. (1990) Vascularity of the 37. Diament J, Keller A, Baer E, Litt M, Arridge. (1972) posterior tibial tendon. J Bone Joint Surg. 72A(6):884–888. Collagen: ultra structure and its relation to mechanical26. Ling SC, Chen CF, Wan RX. (1990) A study of the blood properties as a function of ageing. Proc R Soc Lond. supply of the supraspinatus tendon. Surg Radiol Anat. 180:293–315. 12(3):161–165. 38. Junqueira LC, Contrapulos. (1977) EM of collagen and27. Vascularisation of the long ﬂexor tendon. Okajimas Folia cross striations. In: Junqueira LC, et al., eds. Basic Histol- Anat Jpn. 70(6):285–293. ogy. 2nd ed. Los Altos, CA: Lange.28. Barfred T. (1971) Experimental rupture of the Achilles 39. Vailais AC, Vailais JC. (1994) Physical activity and connec- tendon. Acta Orthop Scand. 42:528–543. tive tissue in physical activity, ﬁtness and health. Hum Kinet.29. Cummings JE, Anson JB, Carr WB, Wright RR, Houser 372–376. DWE. (1946) The structure of the calcaneal tendon (of 40. Kraushaar B, Nirschl R. (1999) Tendinosis of the elbow: Achilles) in relation to orthopedic surgery with additional clinical features and ﬁndings of histological, immuno-histo- observations on the plantaris muscle. Surg Gynecol Obstet. chemical and electron microscopy studies. J Bone Joint 83:107–116. Surg. (Am) 259–278.30. Kvist M. (1994) Achilles tendon injuries in athletes. Sports 41. O’Brien M. (1997) Structure and metabolism of tendons. Med. 18:173–201. Scand J Med Sci Sports. 7:55–61.31. Kastelic J, Galeski A, Baer E. (1978) The multi-composite 42. Scott JE. (1988) Proteoglycan-ﬁbrillar collagen interactions. structure of tendon. Connect Tissue Res. 6:11–23. J Biochem. 252:313–323.32. Borynsenko M, Beringer T. (1989) Functional Histology. 3rd 43. Vogel KG, Sandy JD, Pogany G, Robbins JR. (1994) Aggre- ed. Boston: Little, Brown and Company; 105–112. can in bovine tendon. Matrix Biol. 14(2):171–179.33. Amiel D, Billings E, Akeson WH. (1990) Ligament struc- 44. Vogel KG, Hernandez DJ. (1992) The effects of transform- ture, chemistry, and physiology. In: Daniel D, ed. Knee Lig- ing growth factor-Beta and serum on proteoglycan synthe- aments: Structure, Function, Injury, and Repair. New York: sis by tendon ﬁbrocartilage. Eur J Cell Biol. Dec, 59 Raven Press; 77–91. (2):304–13.34. Khan KM, Cook JL, Bonar F, Harcourt P, Astrom M. (1999) 45. Viidik A. (1973) Functional properties of collagenous Histopathology of common tendinopathies. Sports Med. tissues. Rev Connect Tissue Res. 6:127–215. 27(6):393–408.35. Butler DL, Grood ES, Noyes FR, Zernucke RF. (1978) Bio- mechanics of ligaments and tendons. Exerc Sports Sci Rev. 6:125–182.
2Mechanical Properties of TendonsConstantinos N. Maganaris and Marco V. NariciThe primary role of tendons is to transmit contractile curve. Region I is the initial concave portion of the curve,forces to the skeleton to generate joint movement. In in which stiffness gradually increases; it is referred to asdoing so, however, tendons do not behave as rigid bodies. the tendon “toe” region. Loads within the toe regionIn this chapter, the mechanical behavior of tendons and elongate the tendon by reducing the crimp angle of theits major determinants and implications are reviewed. collagen ﬁbers at rest, but they do not cause further ﬁber stretching. Hence, loading within the toe region does not exceed the tendon elastic limit, and subsequent unload-In Vitro Measurements ing restores the tendon to its initial length. Further elon- gation brings the tendon into the “linear” Region II, inMost of our knowledge of the mechanical properties of which stiffness remains constant as a function of elonga-tendons comes from isolated material testing. Two tion. In this region, elongation is the result of stretchingmethods have traditionally been used in biomechanics imposed in the already aligned ﬁbers by the load imposedinvestigations: 1) The free-vibration method, which is in the preceding toe region. At the end point of thisbased on quantifying the decay in oscillation amplitude region, some ﬁbers start to fail. Thus, A) the tendon stiff-that takes place after a transient load is applied to a spec- ness begins to drop; and B) unloading from this pointimen [1–3]; and 2) tensile testing methodologies, in which does not restore the tendon’s initial length. Elongationthe specimen is stretched by an external force while both beyond the linear region brings the tendon into Regionthe specimen deformation and the applied force are III, where additional ﬁber failure occurs in an unpre-recorded [2,4–6]. The latter methodology seems to be dictable fashion. Further elongation brings the tendonpreferable, mostly because it is considered to mimic ade- into Region IV, where complete failure occursquately the way that loading is imposed on tendons in [4,5,15–18].real life [7–14]. Although Regions I, II, III, and IV are apparent in A tensile testing machine is composed of an oscillat- tendon force-deformation curves during elongation-to-ing actuator and a load cell (see Figure 2-1). The tendon failure conditions, the shape of the curves obtainedspecimen studied is gripped by two clamps, a static one differs between specimens. These differences can bemounted on the load cell and a moving one mounted on accounted for to a great extent by interspecimen dimen-the actuator. The actuator is then set to motion while the sional differences. For example, tendons of equal lengthsload cell records the tension associated with the stretch- but different cross-sectional areas exhibit differenting applied. The tensile deformation of the specimen is force-deformation properties, and thicker tendons aretaken from the displacement of the actuator, in which stiffer. Similarly, different force-deformation curves arecase the deformation of the whole specimen is quantiﬁed, obtained from tendons of equal cross-sectional areas butor by means of an extensometer, in which case deforma- different initial lengths, in which case shorter tendons aretion measurements are taken over a restricted region of stiffer .the whole specimen. To account for interspecimen dimensional differences, A typical force-deformation plot of an isolated tendon tendon force is reduced to stress (MPa) by normalizationis shown in Figure 2-2. Generally, in force-deformation to the tendon cross-sectional area, and tendon deforma-curves, slopes relate to stiffness (N/mm), and areas to tion is reduced to strain (%) by normalization to theenergy (J). In elongation-to-failure conditions, 4 different tendon original length. The tendon stress-strain curve isregions can be identiﬁed in the tendon force-deformation similar in shape to the force-deformation curve, but it14
2. Mechanical Properties of Tendons 15 If a tendon is subjected to a tensile load, the tendon does not behave perfectly elastically, even if the load applied is less than that required to cause failure. This is because the tendon collagen ﬁbers and interﬁber matrix load cell possess viscous properties [20,21]. Due to the presence of viscosity, the entire tendon exhibits force-relaxation, clamp creep, and mechanical hysteresis [2,4,5,8,15,16,22]. Force-relaxation means that the force required to cause a given elongation decreases over time. The decrease in tendon specimen force follows a predictable curvilinear pattern until a steady-state value is achieved (see Figure 2-3). Creep is the analogous phenomenon under constant-force condi- clamp tions. In this case, deformation increases over time curvi- displacement linearly until a steady state value is reached. In both force-relaxation and creep, the decrease in magnitude of force applicationFigure 2-1. Diagram of an apparatus for tendon tensile testing. Forcereﬂects the intrinsic material properties rather than thestructural properties of the specimen. The most common material variables taken from a Time Astress-strain curve under elongation-to-failure conditionsare Young’s modulus (GPa), ultimate stress (MPa), ulti-mate strain (%), and toughness (J/kg). Young’s modulusis the product of stiffness multiplied by the original force applicationlength-to-cross-sectional area ratio of the specimen. DeformationExperiments on several tendons indicate that the Young’smodulus reaches the level of 1 to 2 GPa at stressesexceeding 30 MPa [5,11,12,19]. Ultimate tendon stress(i.e., stress at failure) values in the range of 50 to 100 MPaare generally reported [5,11,12,17]. Ultimate tendon Time Bstrain (i.e., strain at failure) values of 4% to 10% havebeen reported [5,16,17]. The tendon toughness (i.e., workdone on the tendon until failure) values reported are in hysteresisthe range of 1000 to 4500 J/kg . Force I II III IV Deformation C Figure 2-3. (A) Typical force-relaxation curve in a tendon. The Force force required to cause a given deformation decreases over time. (B) Typical creep curve in a tendon. The deformation caused by a given force increases over time. (C) Typical mechan- ical hysteresis in a tendon. The arrows indicate loading and Deformation unloading directions during a test with a tensile load within the elastic limit of the tendon. The area of the loop betweenFigure 2-2. Typical force-elongation curve of a tendon pulled the loading and unloading curves relative to that underneathby a load exceeding the tendon elastic limit. I, toe region; II, the loading curve represents the fraction of strain energy lostlinear region; III and IV, failure regions. as heat by the tendon viscous damping.
16 C.N. Maganaris and M.V. Naricithe variable studied reﬂects the viscous component of the collagen turnover and reducible cross-linking, decreasedtendon, and the steady-state values reﬂect the elastic com- glycosaminoglycan and water content, and increasedponent of the tendon. The presence of mechanical hys- nonuniform orientation of collagen ﬁbrils [5,10,17,18,teresis is retrieved in load-deformation plots during 31–33].loading and subsequent unloading of the specimen[2,6,8,12]. Larger tendon deformations are taken during Physical Activityrecoil than stretch at given loads, yielding a loop (the hys-teresis loop) between the curves in the loading and Most of the studies report that long-term physicalunloading directions (see Figure 2-3).The area of the loop activity improves the tensile mechanical properties ofrepresents the amount of strain energy lost as heat upon tendons and yields opposite effects compared with disuserecoil due to the viscous component, and it is usually [5,9,10,30,34]. Increases in stiffness, ultimate strength, andexpressed in relative terms (%) with respect to the total energy-to-failure in response to exercise training havework performed on the tendon during stretching. been reported. Dimensional changes (i.e., hypertrophy)Mechanical hysteresis values in the range of 5% to 25% may partly account for these changes. Increases inhave been reported, with most values concentrated ultimate stress and strain, however, indicate that thearound the value of 10% [7,8,11,12,19]. The proportion of improvement of mechanical properties is also associatedstrain energy input recovered by elastic recoil is the con- with training-induced changes in the tendon intrinsicverse of mechanical hysteresis, and is known as rebound material properties. Such biochemical and structuralresilience.This variable is, therefore, an index of the mate- changes include increased glycosaminoglycan content,rial potential for elastic energy recovery. decreased collagen, reducible cross-linking, and increased Several factors may account for differences in the alignment of collagen ﬁbers [5,10,17,18,31–33].material properties of tendons. Some differences can beattributed to interstudy methodological differences in A)tendon gripping (conventional clamps, Cryo Jaw clamps, Anatomical Siteor use of cyanoacrylate adhesive [2,6,8,11,12,23]; B) Since chronic physical activity enhances the mechanicaltendon deformation measurement (actuator-based properties of tendons, it would be reasonable to suggestmeasurements, extensometer-based measurements, or that tendons located at anatomical sites that allow high-noncontact optical methodologies [2,6,9,11,24]; and C) level and frequent loading may have enhanced proper-tendon cross-sectional area measurement (gravimetry- ties as compared with tendons loaded by low-level forces.based measurements, micrometry-based measurements, Examples of tendons that are frequently loaded by high-or mass- and density-based estimations [9,11,25]. Some tensile loads are the tendons of the ankle plantarﬂexorstudies have shown that the status of the specimen and digital ﬂexor muscles. These tendons are loaded bystudied (e.g., preserved or fresh) and the environmental the ground impact forces during terrestrial locomotion.conditions during testing may also affect the mechanical At the other end of the spectrum are the tendons ofresponse of collagenous tissue [5,26–28], thus accounting the ankle dorsiﬂexor and digital extensor muscles. Thesefor the above variations. tendons are physiologically loaded primarily by the in- Studies on the effect of several other factors on the series muscles that contract to enable joint displacement.mechanical properties of tendinous tissue have been per- Some experimental results indicate that the location andformed. The major of these factors are discussed below. functional role of a tendon may be associated with the tendon mechanical response [12,35], but more recent studies stand in opposition with the above notion [19,36].DisuseTo determine the effects of disuse on tendinous tissue Agingproperties, 3 limb immobilization models have tradition-ally been employed. In most experiments, the joint is Several studies have shown that aging affects the prop-ﬁxed at a certain position for a prolonged period of time. erties of tendinous tissue [4,5,12,15,18,37,38]. However,Using the specimens of the contralateral, nonimmobi- some studies have shown that aging may result in intrin-lized limb as controls, postintervention comparisons are sically stiffer, stronger, and more resilient tendons [12,13],then made [5,10,29,30]. Limb suspension and denervation while other studies have challenged these results [37–40].models have also been used [29,31]. Most studies show This inconsistency may be partly accounted for by dif-that immobilization results in decreased stiffness, ulti- ferences in the initial age examined. In some studies,mate strength and energy-to-failure. These changes are specimens from very young subjects have been usedattributed to specimen atrophy and changes in the spec- [12,35,41]. On such occasions, changes in tissue proper-imen material properties. Disuse-induced changes in ties reﬂect changes occurring as a function of maturation,intrinsic material properties are associated with increased which may mask an actual aging effect.
2. Mechanical Properties of Tendons 17Steroids distal bone proximal boneCorticosteroids have frequently been used for the treat- muscle tendonment of articular inﬂammations. Intra-articular and intra-collagenous injections of corticosteroid may reduce thestiffness, ultimate stress, and energy-to-failure of collage-nous tissue, even after short-term administration [42–44]. markers stimulationThese results indicate that steroids may predispose the load celluser to tendon injuries. Furthermore, using steroids mayalso impair the tendon healing process after an injury Figure 2-4. Experimental set-up to measure the mechanical. properties of a tendon in situ. The muscle-tendon complex is intact. The proximal and distal bones are clamped. Loading is imposed by stimulation-induced muscle contraction. The resul- tant forces are measured by a load cell placed in series withIn Situ and In Vivo Protocols the muscle-tendon complex. The resultant deformation in the tendon is obtained from off-line analysis of the displacement ofIn vitro–based studies have made it clear that tendons markers attached on the tendon.do not behave as rigid elements. Reference, however, tomechanical properties of in vitro material when inter-preting in vivo function should be treated with caution. the development of a noninvasive method for assessingAlthough frequencies met in physiological locomotion the mechanical properties of human tendons in vivo.have often been used in in vitro tensile tests, three impor- The method is based on real-time, sagittal-plane ultra-tant facts raise doubts as to whether such tests can mimic sound scanning of a reference point along the tendonor predict accurately the tendon mechanical behavior during static contraction of the in-series muscle. The limbunder in vivo loading conditions: 1) Fixing a ﬁbrous struc- is ﬁxed on the load cell of a dynamometer to recordture with clamps is inevitably associated with A) slippage changes in muscle torque during isometric contractionof the outer ﬁbers; and B) stress concentration that may (see Figure 2-5). The tensile forces generated by contrac-result in premature fracture. 2) Many experiments havebeen performed using preserved tendons, which may bhave altered properties [26,27]. 3) Tendon loads withinthe physiological region have traditionally been pre- adicted from the muscle maximal stress potential, whichhas been treated as a constant [12,25,46]. There is exper-imental evidence, however, that maximal muscle stress is e cmuscle-speciﬁc, with ﬁber composition being the major g fdeterminant factor [47–49]. Some of these problems have been circumvented bytesting animal tendons in situ after the animal has been hkilled or anesthetized [50–53]. This has been achievedby surgically releasing the tendon from its surroundingtissues, maintaining the proximal end of the tendon dattached to the in-series muscle, and having the distalbone of the muscle-tendon unit gripped by a clamp inter- ifaced to a load cell.The in situ muscle contracts artiﬁcially jby electrical stimulation and pulls the tendon, whichlengthens as a function of the contractile force applied to Figure 2-5. Experimental set-up to measure the mechanicalits proximal end in a similar fashion to that obtained properties of the human tibialis anterior tendon in vivo. Thewhen the actuator of a tensile machine pulls an isolated limb is ﬁxed on the footplate of a dynamometer. Isometricspecimen (see Figure 2-4). The advantage of such ex- muscle contractions are generated by stimulation, while the re- sultant displacement of the myotendinous junction is recordedperimental protocols is that they allow assessment of in real time using ultrasonography. The load imposed is takenthe tendon and aponeurosis (i.e., intramuscular tendon) from the dynamometer reading. a, dynamometer footplate; b,mechanical properties A) separately, and B) under phys- velcro straps; c, ankle joint; d, tibialis anterior muscle; e, tibialisiological loading levels. Notwithstanding these advan- anterior tendon; f, myotendinous junction; g, ultrasound probe,tages, the above in situ protocols are not applicable to h; percutaneous stimulating electrodes; i, knee joint; j, kneehumans. However, adapting similar principles to those mechanical stop. (Reprinted with permission from Maganarisused under in situ material testing has recently allowed and Paul.)