Review Article. Abstract

Transcription

1 Review Article Acromioclavicular Joint Injuries: Diagnosis and Management Ryan Simovitch, MD Brett Sanders, MD Mehmet Ozbaydar, MD Kyle Lavery, BS Jon J. P. Warner, MD Dr. Simovitch is Attending Surgeon, Palm Beach Orthopaedic Institute, Palm Beach Gardens, FL. Dr. Sanders is Attending Surgeon, Center for Sports Medicine and Orthopaedics, Chattanooga, TN. Dr. Ozbaydar is International Fellow, Harvard Shoulder Service, Massachusetts General Hospital, Boston, MA. Mr. Lavery is Research Assistant, Harvard Shoulder Service, Massachusetts General Hospital. Dr. Warner is Chief, Harvard Shoulder Service, Massachusetts General Hospital. Dr. Warner or a member of his immediate family has received royalties from Zimmer, and research or institutional support from Aircast (DJ), Arthrex, Mitek, Smith & Nephew, and Zimmer. None of the following authors or a member of their immediate families has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article: Dr. Simovitch, Dr. Sanders, Dr. Ozbaydar, and Mr. Lavery. Reprint requests: Dr. Simovitch, Palm Beach Orthopaedic Institute, Suite 500, 3401 PGA Boulevard, Palm Beach Gardens, FL J Am Acad Orthop Surg 2009;17: Copyright 2009 by the American Academy of Orthopaedic Surgeons. Abstract Acromioclavicular joint injuries represent nearly half of all athletic shoulder injuries, often resulting from a fall onto the tip of the shoulder with the arm in adduction. Stability of this joint depends on the integrity of the acromioclavicular ligaments and capsule as well as the coracoclavicular ligaments and the trapezius and deltoid muscles. Along with clinical examination for tenderness and instability, radiographic examination is critical in the evaluation of acromioclavicular joint injuries. Nonsurgical treatment is indicated for type I and II injuries; surgery is almost always recommended for type IV, V, and VI injuries. Management of type III injuries remains controversial, with nonsurgical treatment favored in most instances and reconstruction of the acromioclavicular joint reserved for symptomatic instability. Recommended techniques for stabilization in cases of acute and late symptomatic instability include screw fixation of the coracoid process to the clavicle, coracoacromial ligament transfer, and coracoclavicular ligament reconstruction. Biomechanical studies have demonstrated that anatomic acromioclavicular joint reconstruction is the most effective treatment for persistent instability. Acromioclavicular (AC) joint injuries represent 40% to 50% of athletic shoulder injuries. 1,2 The treatment of AC instability has been an ongoing source of controversy. Long before a three-grade classification of the injury was developed by Tossy et al 3 and Allman 4 in the 1960s and then expanded by Rockwood in 1989, 5 surgeons debated the method and timing of treatment. The greatest source of dispute has been the issue of nonsurgical management versus surgical reconstruction for complete dislocations. In the mid 1970s, most residency program directors in the United States recommended surgical treatment for type III dislocations (ie, loss of contact between the clavicle and the acromion). 6 However, by the early 1990s, 135 of 187 surgeons preferred nonsurgical treatment (72.2%). 7 A series of comparative studies has supported this trend Today, the tendency in management is toward minimal intervention. 13 However, surgical management, most commonly in the form of coracoclavicular (CC) fixation 14 and/or ligament reconstruction, 15,16 is often undertaken after consideration of individual patient demands and injury chronicity. Joint Anatomy and Biomechanics The AC joint is a diarthrodial joint formed between the lateral end of the clavicle and the medial end of the April 2009, Vol 17, No 4 207

2 Acromioclavicular Joint Injuries: Diagnosis and Management Figure 1 The acromioclavicular (AC) joint. Static stabilizers include the AC capsule and the coracoclavicular ligaments, consisting of the trapezoid ligament laterally and the conoid ligament medially. acromion. This joint has a variable inclination between nearly vertical and 50 of obliquity, where the superior edge of the clavicle is more lateral. A fibrocartilaginous disk exists at the AC joint and has been shown to involute with age and disintegrate by age 40 years. 17 Innervation of the AC joint is from the lateral pectoral, axillary, and suprascapular nerves. Multiple studies have confirmed that the clavicle rotates approximately 40 to 45 with full shoulder elevation and abduction. 18 However, the clavicle actually only rotates approximately 5 to 8 relative to the acromion because of synchronous scapuloclavicular (SC) motion. 19 This is the basis for shoulder elevation remaining normal after CC arthrodesis with a lag screw. 20 The AC joint has both static and dynamic stabilizers (Figure 1). The static stabilizers include the AC joint capsule as well as the AC ligaments that reinforce the capsule. Biomechanical investigations have found that these structures predominantly control horizontal motion of the clavicle. 21 Klimkiewicz et al 22 demonstrated in cadaveric specimens that the posterosuperior capsule is essential in preventing excessive posterior translation of the clavicle and therefore that excessive distal clavicle excision can result in increased posterior clavicle translation. 22 The posterior and superior AC ligaments contribute most to the horizontal stability of this joint. Authors of a recent cadaveric study demonstrated that a distal 1 cm clavicle resection results in a 32% increase in posterior translation compared with the intact state. 23 A resection <1 cm also results in increased instability. Renfree and Wright 24 demonstrated that a distal clavicle resection of as little as 2.3 mm in women and 2.6 mm in men could completely release the clavicular insertion of the AC ligaments in some patients. The CC ligaments, which include the conoid ligament medially and the trapezoid ligament laterally, also provide static stabilization and act predominantly as a restraint to vertical translation of the clavicle. 21 Rios et al 25 defined the insertion points of the CC ligaments on the clavicle in a cadaveric study. The normal anatomic range of the distance between the coracoid process and the clavicle (CC interspace) is 1.1 to 1.3 cm. 26 Fukuda et al 21 demonstrated that the conoid ligament restrains vertical displacement to a greater extent than does the trapezoid ligament. They also concluded that the AC joint capsule contributes greater constraint at small loads and that the CC ligaments provide stability with greater displacements. Dynamic stabilization of the AC joint relies on the origin of part of the anterior deltoid muscle from the clavicle and the trapezius muscle through its fascial insertion over the acromion. The role of these muscles in AC joint stability must be appreciated during any AC joint surgical reconstruction. 27 Mechanism of Injury An acute injury to the AC joint can occur through a direct or an indirect mechanism. Direct injury results from a direct force to the acromion with the shoulder adducted, resulting in movement of the acromion inferiorly and medially while the clavicle is stabilized by the sternoclavicular joint ligaments. 25 This mechanism is involved in most injuries and is usually the result of a fall on the superolateral portion of the shoulder. The force results in systematic failure of the stabilizing ligaments with the propagation of increasing force. 20 Failure of the AC ligaments and capsule is followed by failure of the CC ligaments and deltotrapezial fascia. An indirect force can result in the same constellation of injury patterns but is generated by a fall on an outstretched arm or elbow with a superiorly directed force. Although injury to the AC and CC ligaments is the most common pathologic basis for 208 Journal of the American Academy of Orthopaedic Surgeons

3 Ryan Simovitch, MD, et al Table 1 Characterization of Acromioclavicular Joint Injuries by the Rockwood Classification * Type AC Ligaments CC Ligaments Deltopectoral Fascia Radiographic CC Distance Increase Radiographic AC Appearance AC Joint Reducible I Sprained Intact Intact Normal (1.1 to Normal N/A 1.3 cm) II Disrupted Sprained Intact <25% Widened Yes III Disrupted Disrupted Disrupted 25%-100% Widened Yes IV Disrupted Disrupted Disrupted Increased Posterior clavicle No displacement V Disrupted Disrupted Disrupted 100%-300% N/A No VI Disrupted Intact Disrupted Decreased N/A No * The type of AC injury can be discerned based on the pattern of ligament injury, AC joint position on radiographs, and whether the AC joint can be reduced on physical examination. AC = acromioclavicular, CC = coracoclavicular, N/A = not applicable complete AC joint dislocation and instability, AC dislocations can occur in the context of intra- and extraarticular fractures of the base of the coracoid process. Such an injury is considered equivalent to a CC rupture. Epiphyseal separation in children and young adults, as well as distal clavicle fractures, can result in the appearance of AC dislocation on radiographs; these are termed pseudodislocations. 20 Classification The degree of injury to the AC joint is dependent on the amount of energy transferred to the acromion and clavicle and their ligamentous stabilizers. Tossy et al 3 and Allman 4 originally classified AC joint injuries into three grades. This classification system was subsequently modified by Rockwood into six types (Table 1). 5 In type I injury, there is no visible deformity. There may be some swelling and tenderness over the AC joint, but there is no tenderness over the CC interspace. Radiographs appear normal. In type II injury, the distal clavicle is unstable horizontally and can be displaced anteriorly or posteriorly. Although there may be vertical instability, it is usually absent or minimal. In contrast to type I injuries, there is tenderness overlying the CC interspace. Radiographically, the AC joint is disrupted and may be widened, with a slight vertical displacement and concomitant increase in the CC interspace. The distal clavicle is unstable vertically and horizontally in type III injury. Radiographically, the AC joint is dislocated, with the acromion displaced inferiorly relative to the clavicle. There is tenderness in the CC interspace. Pain is typically more severe in type III and higher injuries. The AC joint is reducible by an upward force placed on the ipsilateral elbow. Certain injury patterns mimic, but should be distinguished from, type III AC injuries. These variants include Salter-Harris injuries to the distal clavicular physis resulting from late closure of this physis (in persons aged 18 to 22 years), and AC dislocations associated with intact CC ligaments and a fracture of the coracoid process. AC joint dislocation with the clavicle displaced posteriorly into or through the trapezius muscle identifies a type IV injury. Clinically, the anterior acromion may be prominent. Radiographically, the CC interspace is increased, and posterior translation of the lateral clavicle is seen on an axillary lateral view (Figure 2). The AC joint is irreducible on physical examination. The SC joint and brachial plexus should be evaluated in all AC injuries, but such evaluation is particularly important in type IV injuries, given the reports of bipolar clavicular dislocations (eg, synchronous AC and SC dislocation). In a type V injury, in addition to disruption of all of the stabilizing ligaments (as in a type III or IV injury), the deltoid and trapezius muscles and fascia are more extensively detached from the clavicle. Clinically, the clavicle lies subcutaneously. Occasionally there is so much inferior displacement of the upper extremity that the patient develops symptoms from traction on the brachial plexus. The AC joint is irreducible. The rare type VI injury represents a high-energy variant that usually occurs as a result of hyperabduction and external rotation. The distal clavicle typically comes to rest in a subacromial or subcoracoid position (Figure 3). Because this injury is usu- April 2009, Vol 17, No 4 209

4 Acromioclavicular Joint Injuries: Diagnosis and Management Figure 2 A 33-year-old man presented with shoulder pain and an abrasion over the superior portion of the shoulder following a fall from a motorcycle. A, Anteroposterior radiograph with a 10 to 15 cephalic tilt demonstrating an increase in the coracoclavicular interspace distance (*). B, Axillary lateral radiograph demonstrating posterior displacement (posteriorly directed arrow), thereby distinguishing this as a type IV injury. The acromion (A) and the clavicle (C) are outlined. Figure 3 Three-dimensional computed tomography scan demonstrating the subcoracoid variant of a type VI acromioclavicular dislocation. ally the result of high-energy trauma, it is often accompanied by multiple fractures of the clavicle and ribs. Subcoracoid displacement may be associated with brachial plexus or vascular injury. Diagnosis Clinical Findings Injury to the AC joint should be suspected in any patient who has shoulder trauma with pain in the vicinity of the acromion and clavicle. During the clinical examination, the patient should be in the standing or sitting position without support for the injured arm. The weight of the arm pulling downward often makes the deformity more apparent. On physical examination, findings related to the severity of the injury, such as local swelling, deformity, abrasion, or bruising, may be noted. The distal end of the clavicle is sometimes displaced enough to tent the skin. With AC injury, local tenderness is present on palpation over the AC joint as well as in the CC interspace. AC joint palpation provides guidance related to the degree and direction of the displacement of the clavicle compared with the acromion. The clavicle should also be carefully palpated to detect possible fractures. Disruption of the deltotrapezial fascia should be noted if present. Passive and active motion of the shoulder produces focal pain at the injured AC joint. This pain is often accentuated by abduction and cross-body adduction. The O Brien active compression test for pain over the AC joint may be helpful. 28 The AC joint should next be assessed for stability. Evaluation of stability is difficult in the acute phase because of patient discomfort and guarding. In the subacute phase, after partial resolution of pain, horizontal and vertical instability can be detected. Because an AC joint injury occurs by inferior displacement of the scapula, determining whether the AC joint is reducible is done by stabilizing the clavicle with one hand and placing an upward force under the ipsilateral elbow with the other 210 Journal of the American Academy of Orthopaedic Surgeons

5 Ryan Simovitch, MD, et al hand. A type III AC joint injury can be differentiated from a type IV or V joint injury based on whether the AC joint dislocation can be reduced (Table 1). Once the AC joint is reduced with this maneuver, the examiner can grasp with the index finger and thumb the midshaft of the clavicle and attempt to translate it anteriorly and posteriorly to assess horizontal stability. The sternoclavicular joint should always be examined for an associated anterior dislocation. The neurologic status of the affected extremity should be evaluated in all AC injuries to rule out a brachial plexus injury. Radiographic Evaluation Standardized radiographs are essential to diagnose and classify AC joint injuries. Routine radiographs for AC joint evaluation include a true anteroposterior and an axillary lateral view of the shoulder, as well as Zanca views (10 to 15 cephalic tilt) taken with the patient in an upright position without support of the injured arm. Ideally, comparison views of the uninjured shoulder should be obtained to provide normative information with regard to the CC distance and relative posterior displacement of the distal clavicle. Stress views of the AC joint have been described to differentiate between type II and type III injuries. However, the weighted stress views are costly and uncomfortable for the patient; also, they rarely provide new information to help diagnose an unstable injury. Thus, they are no longer routinely used. Coracoid process fractures should always be suspected when radiographs reveal AC dislocation in the presence of a normal CC distance (1.1 to 1.3 cm). When the coracoid fracture is not obvious on the axillary view, the Stryker notch view will almost always demonstrate it. Treatment Nonsurgical Nonsurgical treatment is uniformly recommended for type I and type II injuries. Although multiple methods for clavicle reduction and immobilization have been described, most authors suggest a period of immobilization in a simple sling or shoulder immobilizer to remove stress from the injured CC and/or AC ligaments. 14,29 Generally, type I injuries can be treated by use of a simple sling for 7 to 10 days or until pain subsides. Type II injuries may require immobilization for as long as 2 weeks for resolution of symptoms. Once the shoulder pain has subsided, an early and gradual rehabilitation program is instituted, with the focus on passive- and active-assisted range of motion (ROM). After symmetric and painless shoulder ROM is achieved, an isometric strengthening program is begun. This is followed by isotonic strengthening with a gradual escalation of strengthening and endurance. Contact sports and heavy lifting should be avoided for 2 to 3 months to allow for ligament healing and avoid converting an incomplete injury to a complete type III injury. Most studies support nonsurgical treatment of type III injuries, although this is controversial. Nonsurgical treatment is well accepted, but mid- and long-term outcomes should not be ignored. Several studies have demonstrated persistent disability in patients treated nonsurgically. Bergfeld et al 30 evaluated United States Naval Academy midshipmen and found that 30% of patients with type I and 42% of patients with type II injuries reported minor symptoms such as clicking and pain with pushups and dips. An additional 9% and 23% of patients with type I and II injuries, respectively, reported severe persistent pain and limitation of activities. Mouhsine et al 31 reported similar results. Of patients with type I and II AC injuries treated nonsurgically, 27% developed chronic AC symptoms at a mean of 26 months after injury and susequently required surgical intervention. Of the patients who chose later surgery, a significant proportion had activity-related pain and residual anteroposterior instability. Bannister et al 9 conducted a randomized, prospective, controlled trial comparing acute surgical treatment of Rockwood type III and V AC joint dislocations using a CC screw versus 2 weeks of immobilization in a sling. The same rehabilitation protocol was used in both groups. Patients were stratified according to the severity of AC joint displacement. Patients with AC displacement <2 cm fared better with nonsurgical treatment. Among only those patients with severe AC dislocation ( 2 cm displacement), 20% of the nonsurgical group had good or excellent results, while 70% of the surgically treated patients had good or excellent results. The authors concluded that nonsurgical treatment is appropriate for low-grade injuries but surgical treatment can be considered for young patients with severe injuries. The most concerning sequelae of lower-energy injuries are the late development of AC joint osteoarthritis, exhibited in nearly 50% of patients with AC injury, 30 and chronic instability. Current treatment options include arthroscopic distal clavicle resection, 32 involving the removal of as little as 4 to 8 mm of bone, and open distal clavicle excision. However, no published prospective, randomized studies exist that compare both arthroscopic and open methods. A meta-analysis by Phillips et al 12 demonstrated that type III injuries have similar overall satisfactory out- April 2009, Vol 17, No 4 211

6 Acromioclavicular Joint Injuries: Diagnosis and Management Table 2 Surgical Techniques for the Management of Acromioclavicular Joint Dislocation Technique Method Features Primary acromioclavicular joint fixation Fixation between the coracoid process and clavicle Kirschner wires, hook-plate CC screw, suture loop Limited dissection Risk of pin migration Reserved for acute injury Screw requires staged removal Suture loop can be inserted arthroscopically Reserved for acute injuries Suture loop can cut through the clavicle or coracoid base Dynamic muscle transfer Coracoid transfer Abandoned due to inferior long-term results Ligament reconstruction Autogenous hamstring, anterior tibialis allograft, CA ligament transfer Attempts to anatomically reestablish CC ligament anatomy Provides biologic scaffold for revascularization Appropriate for acute and chronic reconstructions CA = coracoacromial, CC = coracoclavicular come rates whether managed nonsurgically or surgically. Patients treated nonsurgically returned to work and preinjury activities sooner and had more nearly normal strength and ROM at follow-up. Patients treated surgically had a higher complication rate. However, there was no standardization of surgical technique or nonsurgical treatment method among the studies reviewed. Wojtys and Nelson 33 reviewed 22 patients with type III AC dislocations. They determined that laborers and athletes can recover adequate strength and endurance with nonsurgical treatment of type III injuries despite a slight decrease in both measures. However, they expressed concern that the decrease in endurance and strength, despite not being statistically significant, coupled with persistent pain and soreness, may be worth considering in developing a treatment plan for patients who depend on high levels of shoulder strength or endurance for a job or sport. Multiple other studies have demonstrated successful short- and mid-term clinical outcomes in patients with type III injuries treated nonsurgically Proper and adequate rehabilitation may be critical to successful nonsurgical treatment of type III AC injuries. Glick et al 37 evaluated 35 unreduced AC dislocations that were treated nonsurgically in a professional and competitive recreational athletic population. None of the patients was disabled. In addition, none of the patients who had a supervised rehabilitation program had reports of pain. These authors concluded that the predominant reason for persistent pain and disability after a type III AC injury treated nonsurgically was inadequate rehabilitation. This conclusion is supported by the assertion by Gurd 38 that the shoulder can function normally despite an absent clavicle as long as the shoulder girdle muscles are strengthened and maintained. Thus, a structured active rehabilitation program that focuses on gaining strength of shoulder girdle muscles, including the deltoid, trapezius, sternocleidomastoid, and subclavius, as well as the rotator cuff and periscapular stabilizers, is indicated for patients who are treated nonsurgically. Although there is no conclusive evidence of it in the literature, a relative indication for AC surgical stabilization after an acute type III injury may be young age or a job or sport that places high demands on the shoulder. In addition, chronic symptoms of instability and pain following a type III AC dislocation may warrant delayed surgical reconstruction. Recent reports of successful nonsurgical treatment of type III AC injuries in professional baseball players have challenged the relative indication of sport or job type in selecting a management protocol. 39 Despite the success of nonsurgical treatment, late symptomatic sequelae can develop. Occasionally, distal clavicle osteolysis or a painful AC disk injury, akin to a knee meniscal tear, occurs acutely, or AC arthritis develops over time. When symptomatic, these conditions can be treated by steroid injection or distal clavicle resection with intra-articular disk excision, as described by Mumford, 40 combined with a ligamentous reconstruction to establish stability. Surgical The case can be made for surgical management of some acute type III AC joint dislocations. However, acute type IV, V, and VI AC joint dislocations all require surgical intervention, although there is no consensus on which technique to use. Surgical techniques can be categorized into three main groups (Table 2), although all have the goal of obtaining and retaining anatomic AC joint reduction. 212 Journal of the American Academy of Orthopaedic Surgeons

7 Ryan Simovitch, MD, et al Primary Fixation of the Acromioclavicular Joint Historically, primary fixation of the AC joint consisted of pin fixation using either smooth or threaded Kirschner wires after a closed or open reduction. However, this method has been abandoned, given the rare but catastrophic occurrence of pin migration; the pins have been found in the heart, lungs, and great vessels. 41,42 An alternative technique commonly used in Europe is the hookplate. After open reduction of the AC joint, the lateral end of the plate is inserted deep to the acromion and levered down on the clavicle to reduce the joint. It is secured to the clavicle with bicortical screws. Some surgeons combine hook-plate fixation with ligament reconstruction, 43 while others use the hook-plate alone. 44 The plate is removed at 8 weeks. This technique has been successful in treating AC dislocations and distal clavicle fractures. 43 Sim et al, 44 however, reported that 8 of 16 patients experienced complications (one bent plate, one plate dislocation, six infections). Fixation Between the Coracoid Process and Clavicle Bosworth 45 popularized the use of a screw for CC fixation. The use of screws and suture loops has been described alone and in combination with ligament reconstruction. Rockwood et al 20 recommended that a CC screw (Figure 4) be combined with ligament reconstruction in acute cases of AC dislocation. Through recent arthroscopic advances, CC fixation can now be accomplished by less invasive means. Placement of synthetic loops between the coracoid process and clavicle have been described by several authors, most often in combination with a form of CC ligament reconstruction. The main advantage of Figure 4 A, Intraoperative photograph demonstrating percutaneous placement of a coracoclavicular (CC) screw under fluoroscopic guidance. The screw is placed to reduce the CC interspace and anchor the clavicle to the coracoid process during soft-tissue healing, as demonstrated in an intraoperative anteroposterior radiograph (B). this technique is that it does not require staged removal as do CC screws. Absorbable and nonabsorbable materials can be used. Morrison and Lemos 46 demonstrated the importance of accurate positioning of a synthetic loop to allow for the most anatomic reduction possible. Potential complications of this technique include suture cutout. Good results have been reported with 6-mm polytetrafluoroethylene surgical tape as a secondary fixation combined with coracoacromial (CA) ligament transfer. 46 However, cases of aseptic foreign-body reaction and clavicle osteolysis that resulted in failure have been reported. 47,48 Coracoid Process Transfer Transfer of the coracoid process was first described by Dewar and Barrington. 49 It has been used for the treatment of acute and chronic injuries, with better results in younger patients. However, reports of residual joint aching led to the procedure s being abandoned because of poor long-term outcomes. 50 Ligament Reconstruction Weaver and Dunn 15 first described the use of the native CA ligament to reestablish AC joint stability (Figure 5). This technique has since been modified to include resection of the distal clavicle to avoid late degenerative changes at the AC joint. The CA ligament is detached from the deep surface of the acromion with or without a chip of bone and is then transferred to the clavicle. This construct can be augmented with a suture loop that provides protection while the reconstructed ligament heals. Recently, Lafosse et al 51 described an all-arthroscopic technique for CA ligament transfer in the setting of acute or chronic AC dislocations. An alternative technique for ligament reconstruction is the use of a semitendinosus tendon autograft. This technique is combined with re- April 2009, Vol 17, No 4 213

8 Acromioclavicular Joint Injuries: Diagnosis and Management Figure 5 The modified Weaver-Dunn procedure. The coracoacromial ligament (A) is transferred to the clavicle to substitute for the ruptured coracoclavicular ligaments. A suture loop (B) can be used for augmentation. section of the distal clavicle. Jones et al 52 described the use of a looped semitendinosus graft around the coracoid process and clavicle in a revision AC joint reconstruction. Mazzocca et al 16 modified this technique, incorporating a doubled semitendinosus graft inserted in a coracoid bone tunnel and secured in two separate clavicle bone tunnels. Interference screws were used to approximate the anatomic location of the trapezoid ligaments. Biomechanical testing of this construct has been favorable. We have used anterior tibialis tendon allograft in reconstruction (Figure 6). Good clinical results have been achieved using hamstring tendons to reconstruct high-grade AC separations. 53 However, future studies must examine the extent to which two drill holes weaken the clavicle and how tunnel expansion 54 influences the biomechanical and clinical outcomes of this treatment. We favor the use of a doubled semitendinosus autograft or tibialis anterior allograft looped around the Figure 6 Intraoperative photographs of a procedure to anatomically reconstruct the coracoclavicular ligaments through bone tunnels in the clavicle (C) using a semitendinosus allograft. A, A suture passer is passed from medial to lateral around the coracoid tip (*) and used to retrieve the anterior tibialis allograft around the coracoid process. B, The graft ends are pulled through two bone tunnels (arrows) in the clavicle to approximate the pull of the conoid and trapezoid ligaments. 214 Journal of the American Academy of Orthopaedic Surgeons

9 Ryan Simovitch, MD, et al coracoid process or potted in a coracoid bone socket. The graft is fixed with interference screws through two separate clavicle bone tunnels that approximate the normal anatomic location and orientation of the CC ligaments. We believe that this best replicates the injured anatomy. Biomechanical studies have demonstrated the superiority of this construct. We routinely resect the distal clavicle at the index operation. However, recent interest in the additional AC stability conferred by not resecting the distal clavicle and the clinical ramifications of doing so may be compelling. A possible shift in the treatment paradigm awaits further prospective clinical and well-designed biomechanical trials. Distal Clavicle Resection There has been some interest in primary or delayed excision of the distal clavicle (ie, Mumford procedure) in a symptomatic AC joint after dislocation. Although there has been success with this technique done open as well as arthroscopically, it must be reserved for patients in whom the CC ligaments are intact and there is no concomitant instability. When horizontal or vertical instability exists, results are compromised because this technique does not address instability and may accentuate it. 23,24,55 Biomechanical Considerations The goal of AC joint reconstruction, when indicated, is to achieve an anatomic reconstruction that best restores both the restraint to vertical as well as anteroposterior translation of the clavicle at the AC joint. Multiple studies have examined the biomechanical results of various reconstructive techniques (Table 3), and several conclusions can be drawn from them. 16,56-63 A study by Jari et al 56 suggests that surgical techniques that preserve the articulating surface of the clavicle at the AC joint lead to lower joint contact forces and are preferred. CA ligament transfer, which has long been the most popular method for treating complete chronic AC dislocations, has just 25% of the strength of the intact CC complex and allows far greater primary and coupled translation than does the intact AC joint. 63 Augmentation of this repair/reconstruction with various suture loops and suture anchors can increase the construct strength and ultimate load to failure while reducing primary and coupled translation. However, these constructs are biomechanically inferior to the native intact AC and CC capsuloligamentous unit. Screw fixation may also be considered. Although the bicortical Rockwood screw has the highest tensile strength and stiffness with the lowest elasticity, these properties may also be responsible for its high complication rate, including screw breakage and pullout. 56 In a recent study of 42 cadaveric specimens, Mazzocca et al 16 compared the stability conferred by three AC joint reconstruction techniques: CA ligament transfer and distal clavicle resection with suture loop augmentation, anatomic reconstruction with a doublebundled semitendinosus graft inserted into a coracoid bone socket, and arthroscopic reconstruction with suture and titanium screws (ie, suture loop and CC screw fixation). Only the anatomic CC reconstruction with tendon graft and suture augmentation provided anterior, posterior, and superior stability not statistically different from the intact state. The authors also recognized that a graft offers the advantage of a scaffold for revascularization compared with the arthroscopic suture loop, which does not and which can fatigue and fail over time with cyclic loading. Costic et al 57 demonstrated that the dislocated clavicle can lose up to 40% of its stiffness, which suggests that previous biomechanical models may be flawed, depending on how the clavicle and scapula were potted for testing. Complications Complications can occur as a result of nonsurgical and surgical treatment of AC joint dislocations. The most common complications associated with nonsurgical treatment are development of late arthrosis and persistent instability. Distal clavicle osteolysis also has been described. Complications following surgical treatment of AC joint dislocations are related to the technique chosen. Hardware failure and migration resulting in injury to the great vessels have been described. In addition, aseptic foreignbody reaction or infection may occur after the use of implants and synthetic suture. Depending on the choice of technique for surgical reconstruction, early or late fracture of the coracoid process or clavicle has been reported. In addition, any technique that passes a graft or synthetic material medial to the coracoid process poses a potential risk to the brachial plexus and axillary artery. Although reconstructive success is clinically predicated on alleviating pain and establishing AC stability, persistent pain and recurrent instability have been reported with all of the described techniques. Postoperative Rehabilitation After CC fixation with a metallic screw or suture construct for acute AC injuries, the shoulder is immobilized in a simple sling and cold therapy device. At 2 weeks, active and April 2009, Vol 17, No 4 215

10 Acromioclavicular Joint Injuries: Diagnosis and Management Table 3 Biomechanical Comparison of Acromioclavicular Joint Reconstruction Techniques Study Methodology Results Mazzocca et al 16 Jari et al 56 Costic et al 57 Deshmukh et al 58 Baker et al 59 Lee et al 63 Wilson et al 60 Biomechanical testing of 42 cadaveric shoulders for vertical and AP translation of the AC joint after three reconstructive techniques: modified Weaver-Dunn procedure, arthroscopic suture fixation, and anatomic CC reconstruction with semitendinosus graft Biomechanical comparison of the CA ligament transfer, CC sling, and Rockwood screw Biomechanical comparison of an intact AC joint with anatomic reconstruction using a semitendinosus graft Biomechanical comparison of the Weaver-Dunn procedure alone with the Weaver-Dunn procedure augmented by suture cerclage or suture anchor fixation Biomechanical examination of varied placement of the drill hole for loop fixation of the clavicle after AC dislocation in 14 cadaveric shoulders Biomechanical comparison of CA ligament transfer, Mersilene tape reconstruction, and ligament reconstruction with semitendinosus, gracilis, and long toe extensors Biomechanical comparison of Weaver-Dunn reconstruction alone compared with Weaver- Dunn augmented with CC suture anchor fixation All three techniques establish comparable vertical stability and load to failure Anatomic CC reconstruction that reconstructs both conoid and trapezoid ligaments allows the least AP translation of the three techniques and best approximates the intact state Compared with the intact AC joint state, anterior, posterior, and superior translation increased 110%, 100%, and 360%, respectively, after the CA ligament transfer. With the CC sling, anterior translation increased by 110% and posterior translation by 330%. Primary translation was reduced with the Rockwood screw. A significant decrease in posterior translation was noted (60%) compared with the intact joint. Similar findings were noted for coupled translations (ie, AP). The Rockwood screw was the most rigid construct and was found to experience forces 30% and 170% greater than the intact AC joint in response to anterior and posterior loads, respectively. Stiffness and ultimate load to failure of the intact CC ligament state was significantly greater than for the anatomic reconstruction complex (P 0.05) Stiffness of the anatomic reconstruction complex (23.4 N/mm) was 40% of the stiffness of the intact CC complex (60.8 N/mm) Anatomic reconstruction complex had clinically insignificant elongation (<3 mm) after cyclic loading There is a 40% decrease in bending stiffness of the clavicle after dislocation that contributes to diminished stiffness of anatomic reconstruction Average load to failure of the Weaver-Dunn procedure alone was 177 N compared with 319 N for augmentation with a variety of suture anchors or suture cerclage The only failure during the 1,000-cycle fatigue test occurred with an unaugmented Weaver-Dunn procedure The augmented Weaver-Dunn technique significantly better approximated the intact state of the AC joint for superior and AP laxity compared with the unaugmented Weaver-Dunn (P = 0.05) The mean superior and AP laxity for the augmented Weaver-Dunn technique was 7.6 mm and 26.9 mm, respectively, versus 3.1 mm and 8.8 mm, respectively, for the intact state. The authors found no significant difference in biomechanical properties depending on suture anchor choice for the augmentation technique A more anterior drill hole in the clavicle allows more joint congruity and less anterior displacement of the clavicle with anterior loading No method of loop fixation approached normal AC joint congruity CA ligament transfer was the weakest construct Mersilene tape reconstruction provided superior initial fixation strength compared with CA ligament transfer Ultimate load to failure was equivalent among semitendinosus, gracilis, and long toe extensor grafts AC joints were significantly more mobile after Weaver-Dunn reconstruction alone compared with the intact state (P < 0.005). However, with suture anchor augmentation, the reconstructed state approached the intact state in terms of anterior, posterior, and superior translation. AC = acromioclavicular, AP = anteroposterior, CA = coracoacromial, CC = coracoclavicular 216 Journal of the American Academy of Orthopaedic Surgeons

11 Ryan Simovitch, MD, et al Table 3 (continued) Biomechanical Comparison of Acromioclavicular Joint Reconstruction Techniques Study Methodology Results Breslow et al 61 Harris et al 62 Biomechanical comparison of suture and suture anchors for the loop method of stabilizing the dislocated AC joint Biomechanical comparison of tensile strength, tensile stiffness, and elongation at failure with five techniques in 19 cadaveric shoulders: CA transfer, CC sling with 8-mm polyester vascular prosthesis, 2 CC suture anchors, unicortical Bosworth screw, and bicortical Bosworth screw After 10,000 cycles, laxity of the construct with suture loop was 1.32 ± 0.59 mm compared with 1.33 ± 0.94 mm for the suture anchor loop. The differences were not statistically significant. CA ligament transfer demonstrated the weakest reconstruction initially CC slings had high tensile strength but were very elastic and had the highest elongation at failure. CC screws provided the highest tensile strength and stiffness as well as least elasticity, but only with bicortical coracoid purchase AC = acromioclavicular, AP = anteroposterior, CA = coracoacromial, CC = coracoclavicular passive ROM is initiated and restricted to beneath the shoulder level. Doing so allows the patient to begin activities of daily living while avoiding lifting anything heavier than 5 pounds. Once the screw is removed at 2 to 3 months, full active and passive ROM is encouraged, with strengthening limited to light resistance for 6 to 8 weeks. Once full ROM and strength are obtained, return to athletic competition is permitted. After AC joint reconstruction with an autogenous or allograft ligament reconstruction (eg, semitendinosus graft, modified Weaver-Dunn), the arm is maintained in a simple sling. At 2 weeks, pendulum exercises are initiated, followed by light activities of daily living at 4 weeks. With further graft maturation at approximately 8 weeks, active and passive ROM is encouraged with a therapist, and light resistance can be initiated after 3 months. Once full ROM and strength are obtained, return to athletic competition or manual labor is permitted. Summary Significant recent advances have been made in the approach to AC joint injury. There is a consensus that type I and II AC joint injuries should be treated nonsurgically, while acute type IV, V, and VI injuries should be treated surgically. The correct algorithm for treating type III injuries is not known; most studies do not show a significant difference in the clinical outcome between nonsurgically and surgically treated patients. Although it has not been sufficiently demonstrated, it may be that a subset of overhead athletes and heavy laborers would benefit from surgical reconstruction of type III injuries. The idea that adequate rehabilitation is critical to a successful outcome of nonsurgical treatment of a type III injury is worthy of attention and further study. It is often the case that nonsurgical care translates into benign neglect, and perhaps inadequate rehabilitation has been responsible for some of the failures related to nonsurgical treatment. References Citation numbers printed in bold type indicate references published within the past 5 years. 1. Kaplan LD, Flanigan DC, Norwig J, Jost P, Bradley J: Prevalence and variance of shoulder injuries in elite collegiate football players. Am J Sports Med 2005; 33: Thorndike AJ: Injuries to the acromioclavicular joint: A plea for conservative treatment. Am J Surg 1942; 55: Tossy JD, Mead MC, Sigmond HM: Acromioclavicular separations: Useful and practical classification for treatment. Clin Orthop Relat Res 1963;28: Allman FL Jr: Fractures and ligamentous injuries of the clavicle and its articulation. J Bone Joint Surg Am 1967; 49: Williams GR Jr, Nguyen VD, Rockwood CA Jr: Classification and radiographic analysis of acromioclavicular dislocations. Appl Radiol 1989;18: Powers JA, Bach PJ: Acromioclavicular separations: Closed or open treatment? Clin Orthop Relat Res 1974;104: Cox JS: Current method of treatment of acromioclavicular joint dislocations. Orthopedics 1992;15: Larsen E, Bjerg-Nielsen A, Christensen P: Conservative or surgical treatment of acromioclavicular dislocation: A prospective, controlled, randomized study. J Bone Joint Surg Am 1986;68: Bannister GC, Wallace WA, Stableforth PG, Hutson MA: The management of acute acromioclavicular dislocation: A randomised prospective controlled trial. J Bone Joint Surg Br 1989;71: Taft TN, Wilson FC, Oglesby JW: April 2009, Vol 17, No 4 217

12 Acromioclavicular Joint Injuries: Diagnosis and Management Dislocation of the acromioclavicular joint: An end-result study. J Bone Joint Surg Am 1987;69: Galpin RD, Hawkins RJ, Grainger RW: A comparative analysis of operative versus nonoperative treatment of grade III acromioclavicular separations. Clin Orthop Relat Res 1985;193: Phillips AM, Smart C, Groom AF: Acromioclavicular dislocation: Conservative or surgical therapy. Clin Orthop Relat Res 1998;353: Bradley JP, Elkousy H: Decision making: Operative versus nonoperative treatment of acromioclavicular joint injuries. Clin Sports Med 2003;22: Lemos MJ: The evaluation and treatment of the injured acromioclavicular joint in athletes. Am J Sports Med 1998;26: Weaver JK, Dunn HK: Treatment of acromioclavicular injuries, especially complete acromioclavicular separation. J Bone Joint Surg Am 1972;54: Mazzocca AD, Santangelo SA, Johnson ST, Rios CG, Dumonski ML, Arciero RA: A biomechanical evaluation of an anatomical coracoclavicular ligament reconstruction. Am J Sports Med 2006; 34: DePalma AF: The role of the discs of the sternoclavicular and acromioclavicular joints. Clin Orthop Relat Res 1959;13: Kennedy JC: Complete dislocation of the acromioclavicular joint: 14 years later. J Trauma 1968;8: Flatow EL: The biomechanics of the acromioclavicular, sternoclavicular, and scapulothoracic joints. Instr Course Lect 1993;42: Rockwood CA Jr, Williams GR Jr, Young DC: Disorders of the acromioclavicular joint, in Rockwood CA, Matsen FA (eds): The Shoulder. Philadelphia, PA: WB Saunders, 1998, pp Fukuda K, Craig EV, An KN, Cofield RH, Chao EY: Biomechanical study of the ligamentous system of the acromioclavicular joint. J Bone Joint Surg Am 1986;68: Klimkiewicz JJ, Williams GR, Sher JS, Karduna A, Des Jardins J, Iannotti JP: The acromioclavicular capsule as a restraint to posterior translation of the clavicle: A biomechanical analysis. J Shoulder Elbow Surg 1999;8: Corteen DP, Teitge RA: Stabilization of the clavicle after distal resection: A biomechanical study. Am J Sports Med 2005;33: Renfree KJ, Wright TW: Anatomy and biomechanics of the acromioclavicular and sternoclavicular joints. Clin Sports Med 2003;22: Rios CG, Arciero RA, Mazzocca AD: Anatomy of the clavicle and coracoid process for reconstruction of the coracoclavicular ligaments. Am J Sports Med 2007;35: Bearden JM, Hughston JC, Whatley GS: Acromioclavicular dislocation: Method of treatment. J Sports Med 1973;1: Lizaur A, Marco L, Cebrian R: Acute dislocation of the acromioclavicular joint: Traumatic anatomy and the importance of deltoid and trapezius. J Bone Joint Surg Br 1994;76: O Brien SJ, Pagnani MJ, Fealy S, McGlynn SR, Wilson JB: The active compression test: A new and effective test for diagnosing labral tears and acromioclavicular joint abnormality. Am J Sports Med 1998;26: Nuber GW, Bowen MK: Acromioclavicular joint injuries and distal clavicle fractures. J Am Acad Orthop Surg 1997;5: Bergfeld JA, Andrish JT, Clancy WG: Evaluation of the acromioclavicular joint following first- and second-degree sprains. Am J Sports Med 1978;6: Mouhsine E, Garofalo R, Crevoisier X, Farron A: Grade I and II acromioclavicular dislocations: Results of conservative treatment. J Shoulder Elbow Surg 2003;12: Nuber GW, Bowen MK: Arthroscopic treatment of acromioclavicular joint injuries and results. Clin Sports Med 2003;22: Wojtys EM, Nelson G: Conservative treatment of grade III acromioclavicular dislocations. Clin Orthop Relat Res 1991;268: Bjerneld H, Hovelius L, Thorling J: Acromio-clavicular separations treated conservatively: A 5-year follow-up study. Acta Orthop Scand 1983;54: Cox JS: The fate of the acromioclavicular joint in athletic injuries. Am J Sports Med 1981;9: Walsh WM, Peterson DA, Shelton G, Neumann RD: Shoulder strength following acromioclavicular injury. Am J Sports Med 1985;13: Glick JM, Milburn LJ, Haggerty JF, Nishimoto D: Dislocated acromioclavicular joint: Follow-up study of 35 unreduced acromioclavicular dislocations. Am J Sports Med 1977;5: Gurd FB: The treatment of complete dislocation of the outer end of the clavicle: An hitherto undescribed operation. Ann Surg 1941;113: McFarland EG, Blivin SJ, Doehring CB, Curl LA, Silberstein C: Treatment of grade III acromioclavicular separations in professional throwing athletes: Results of a survey. Am J Orthop 1997;11: Mumford EB: Acromioclavicular dislocation: A new operative treatment. J Bone Joint Surg Am 1941;23: Norrell H Jr, Llewellyn RC: Migration of a threaded Steinmann pin from an acromioclavicular joint into the spinal canal: A case report. J Bone Joint Surg Am 1965;47: Sethi GK, Scott SM: Subclavian artery laceration due to migration of a Hagie pin. Surgery 1976;80: Faraj AA, Ketzer B: The use of a hookplate in the management of acromioclavicular injuries: Report of ten cases. Acta Orthop Belg 2001;67: Sim E, Schwarz N, Höcker K, Berzlanovich A: Repair of complete acromioclavicular separations using the acromioclavicular-hook plate. Clin Orthop Relat Res 1995;314: Bosworth BM: Acromioclavicular separation: New method of repair. Surg Gynecol Obstet 1941;73: Morrison DS, Lemos MJ: Acromioclavicular separation: Reconstruction using synthetic loop augmentation. Am J Sports Med 1995; 23: Boldin C, Fankhauser F, Ratschek M, Haller-Schober EM: Foreign-body reaction after reconstruction of complete acromioclavicular dislocation using PDS augmentation. J Shoulder Elbow Surg 2004;13: Stewart AM, Ahmad CS: Failure of acromioclavicular reconstruction using Gore-Tex graft due to aseptic foreignbody reaction and clavicle osteolysis: A case report. J Shoulder Elbow Surg 2004;13: Dewar FP, Barrington TW: The treatment of chronic acromio-clavicular dislocation. J Bone Joint Surg Br 1965; 47: Ferris BD, Bhamra M, Paton DF: Coracoid process transfer for acromioclavicular dislocations: A report of 20 cases. Clin Orthop Relat Res 1989; 242: Lafosse L, Baier GP, Leuzinger J: Arthroscopic treatment of acute and chronic acromioclavicular joint dislocation. Arthroscopy 2005;21: Journal of the American Academy of Orthopaedic Surgeons