➢ There is a general consensus that type-I and II acromioclavicular injuries should be treated nonoperatively with limited immobilization and early return to function.
➢ The treatment of type-III acromioclavicular injuries is controversial, and there is fair evidence that favors initial nonoperative treatment.
➢ There is insufficient evidence to recommend primary operative treatment for overhead athletes or manual laborers with acute type-III acromioclavicular joint dislocation.
➢ There is a general consensus that high-grade acromioclavicular injuries (types IV, V, and VI) should be treated operatively.
➢ There is insufficient evidence to recommend the optimum timing of operative repair (early or delayed), the ideal type of operative repair (anatomic or nonanatomic), the type of graft for operative repair (allograft or autograft), and the method of repair (open or arthroscopic) for chronic acromioclavicular joint dislocations.
➢ There is increasing interest in anatomic coracoclavicular ligament reconstruction for the treatment of chronic acromioclavicular joint dislocations, but there is insufficient evidence to recommend this as the operative procedure of choice for the treatment of complete acromioclavicular joint dislocations.
Acromioclavicular joint injury is a common contact sports-related injury in the upper extremity1. These injuries range from a simple sprain of the acromioclavicular ligament to a complete dislocation (complete separation) of the joint, which involves disruption of the acromioclavicular and coracoclavicular ligaments and varying degrees of injury to the deltoid and trapezius muscles and fascia.
Rockwood et al.2 classified acromioclavicular joint injuries on the basis of the number of ligaments involved, the severity of injury, radiographic findings (including the position of the clavicle relative to the acromion), and reducibility of the acromioclavicular joint with shoulder shrugging (Table I). The Upper Extremity Committee of ISAKOS (International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine) recently proposed a subclassification of the original Rockwood type-III dislocation into stable (type-IIIA) and unstable (type-IIIB) variants. The unstable variant (type IIIB) was characterized by the presence of an overriding distal part of the clavicle on the anteroposterior radiograph with cross-arm adduction view. Patients with this variant have substantial scapular dyskinesia that does not respond to a trial of physical therapy (therapy-resistant scapular dyskinesia)3.
The treatment of acromioclavicular joint injuries has been a subject of considerable debate because of a lack of consensus on the indications for operative treatment and the timing and type of operative repair, especially with regard to type-III acromioclavicular joint dislocations. We present an evidence-based critical analysis of the operative and nonoperative treatment of acromioclavicular joint injuries with emphasis on the treatment of type-III acromioclavicular joint dislocations, which are a topic of continued debate. The literature was critically appraised to answer the following queries: (1) What is the optimum treatment (operative or nonoperative) for type-III acromioclavicular joint dislocation? (2) What is the optimum timing of operative repair of acromioclavicular joint dislocation? (3) Which operative technique (anatomic or nonanatomic) is ideal for the treatment of complete acromioclavicular joint dislocation? In addition, evidence-based recommendations for the treatment of acromioclavicular joint injury are also presented in this critical review.
Two investigators conducted a search of the PubMed database on June 3, 2014, by searching the single term acromioclavicular. Search limits were set to the following: English language, humans, and all adult (nineteen years of age and older) (Fig. 1). The focus of this critical analysis was on the treatment of type-III acromioclavicular joint dislocations, and a meta-analysis of the studies that met the inclusion criteria was performed to formulate recommendations for the treatment of acromioclavicular injuries. Studies were included if they compared operative and nonoperative treatment, early and delayed treatment, or anatomic and nonanatomic operative repair as described previously4. Anatomic repairs were defined as techniques that reproduced both the conoid and trapezoid ligaments by means of graft or nonabsorbable suture. Nonanatomic techniques were defined as repairs that reproduced a single coracoclavicular ligament or procedures that involved open reduction and internal fixation with use of implanted materials such as plates, pins, screws, or wires. The exclusion criteria included Level-V evidence, laboratory studies, radiographic studies, biomechanical studies, fractures or revision acromioclavicular joint procedures, meta-analyses, and studies evaluating preliminary results. The studies that were included in this critical analysis were evaluated, and the following data were extracted from each manuscript: study design, level of evidence, number of patients, mean age, mean duration of follow-up, evaluation of anatomic reduction, functional outcome measures, and complications. Additional data on return to work and return to sport were collected from studies comparing operative and nonoperative treatment and studies comparing early and delayed treatment of acromioclavicular injuries. The level of evidence was assigned according to the Levels of Evidence for primary research questions as described in The Journal of Bone & Joint Surgery5. The extracted data were then categorized according to the three question groups being investigated in this critical analysis: (1) operative compared with nonoperative treatment, (2) early compared with delayed treatment, and (3) anatomic compared with nonanatomic repair of acromioclavicular injuries.
In addition, a thorough literature search was performed on the treatment of type-I, II, IV, V, and VI acromioclavicular injuries. Important topics related to the treatment of acromioclavicular joint injuries, including the natural history and pathology of acromioclavicular injuries and understanding of the pathology and biomechanical rationale behind various surgical techniques, were reviewed to present conceptual material in this manuscript.
The search that was conducted on June 3, 2014, resulted in 925 citations (Fig. 1). Of these, 666 studies were excluded on the basis of their titles as they clearly were not related to the purpose of this review. The abstracts of the remaining 259 citations were reviewed, which resulted in an additional 142 exclusions. Of the remaining 117 studies, ninety-five were excluded after review, resulting in a final group of twenty-two studies to be included from the literature. The final group of studies that were evaluated for this critical analysis included fourteen studies comparing operative and nonoperative treatment, five studies comparing early and delayed treatment, and three studies comparing anatomic and nonanatomic operative techniques.
Treatment of Acromioclavicular Joint Injuries
The rationale for the treatment of acromioclavicular joint dislocation has evolved substantially over the last three decades. The goals of treatment are to achieve a pain-free shoulder with full range of motion and strength and to allow a return to the previous level of activity. Historically, acromioclavicular joint injuries have been predominantly treated nonoperatively6. With the recognition that complete dislocations were associated with poor outcome, there was increased interest in operative fixation of severe acromioclavicular joint injuries. Numerous surgical procedures have been described in the literature, but each has its own limitations and no gold standard treatment option is available7. The choice of treatment (operative or nonoperative) is guided by the activity level of the patient, the severity of symptoms, and the grade of injury, which indicates the severity of ligamentous injury. The relative risks and benefits of operative and nonoperative treatment should be taken into account when developing algorithms for the treatment of acromioclavicular joint injuries.
Nonoperative treatment of acromioclavicular joint dislocation does not restore anatomy but is directed toward early mobilization and restoration of function. Consequently, the deformity persists and there is logical concern about acromioclavicular joint arthritis and instability due to persistent dislocation of the acromioclavicular joint. However, complications associated with operative treatment are avoided with nonoperative treatment. Operative treatment, on the other hand, is directed toward restoring anatomy and aims to provide horizontal and vertical stability in an effort to reduce pain and maximize function. It is thought that, by reducing the joint and providing stability, operative treatment will prevent arthritis. However, operative treatment has its own limitations in the form of a surgical scar and complications including but not limited to loss of reduction or failure of fixation; implant-related complications, including failure and migration; clavicular or coracoid fracture; infection; and stiffness.
Operative and Nonoperative Treatment of Type-I and II Acromioclavicular Joint Injuries
There is a general consensus in the literature that low-grade acromioclavicular injuries (types I and II) can be treated nonoperatively. Nonoperative treatment involves initial pain control followed by immediate rehabilitation for scapular control and shoulder range-of-motion and strengthening exercises. Nonoperative treatment of type-I and II injuries has been associated with good short-term results in the majority of patients8-10. Interestingly, anatomic reduction of the acromioclavicular joint has not been shown to correlate with functional outcome in this subset of patients8-10. However, intermediate and long-term outcome studies have demonstrated that the morbidity of lower-grade acromioclavicular injuries is underestimated8-11. Mouhsine et al.9 retrospectively reviewed thirty-three patients with type-I and II acromioclavicular joint injuries and reported that 50% of these patients were asymptomatic after a mean duration of follow-up of 6.3 years. Furthermore, nine patients (27%) were symptomatic enough to undergo subsequent surgery (distal clavicular resection and acromioclavicular ligament reconstruction). Cox8 retrospectively reviewed 165 acromioclavicular joint injuries and demonstrated substantial residual symptoms in 28% and 35% of type-I and type-II injuries, respectively.
Operative and Nonoperative Treatment of Type-III Acromioclavicular Joint Dislocation
The ideal treatment of type-III acromioclavicular joint dislocation is a matter of ongoing debate. Prospective and retrospective cohort analyses of untreated patients with type-III acromioclavicular joint dislocations have helped us understand the natural history of this subtype12-16. Schlegel et al.12 prospectively evaluated a cohort of twenty-five patients with acute, untreated type-III acromioclavicular joint dislocations. The objective assessment at one year demonstrated full painless range of motion in all of the patients who were satisfied with the outcome (n = 16; 80%). Strength testing demonstrated a mean 17% loss of strength, especially in the bench press activity, but demonstrated no significant loss of strength during rotational motion testing. The majority of patients noticed a cosmetic deformity, and 20% of the patients found the deformity unacceptable. Dias et al.14 and Rawes and Dias15 retrospectively followed a cohort of patients with type-III acromioclavicular joint dislocations and reported successful intermediate and long-term results, respectively, after conservative treatment. Glick et al.16 retrospectively evaluated thirty-four male patients with complete acromioclavicular joint dislocations after a mean duration of follow-up of three years (range, one to ten years). A majority of the patients were athletes: nineteen patients were professional football players, and nine patients were competitive or recreational athletes. The patients were managed with an aggressive rehabilitation protocol, and most of the competitive athletes returned to playing their sport within four weeks. At the time of evaluation, all of the professional football players had a painless acromioclavicular joint and full range of shoulder motion. The aforementioned studies provide us with a critical insight into the nonoperative treatment of type-III dislocations12-16. First, incomplete reduction of the acromioclavicular joint dislocation does not necessarily correlate with poor outcome. Second, chronic changes in the dislocated acromioclavicular joint, osteolysis or hypertrophy of the distal part of the clavicle, and calcification of the coracoclavicular ligaments do not necessarily correspond with a poor functional outcome or painful shoulder. Third, the deformity associated with acromioclavicular joint dislocation does not completely reduce with time, although it may reduce in severity in certain cases. Fourth, a proportion of patients do not do well with nonoperative treatment, but they do not constitute the majority.
Operative treatment of a type-III acromioclavicular joint dislocation is recommended in cases of failure of nonoperative treatment. The most common presentation of failure of nonoperative treatment includes persistent pain, scapular dysfunction, and weakness and/or fatigue with active use of the involved upper extremity. It has also been suggested that primary operative intervention should be considered for the treatment of type-III acromioclavicular joint dislocation in throwing and contact athletes and heavy manual laborers, but there is limited evidence to support this recommendation16-19.
Critical Analysis of Operative and Nonoperative Treatment of Type-III Acromioclavicular Joint Dislocation
Of the twenty-two studies that were included in this critical analysis review, fourteen studies compared operative with nonoperative treatment (Table II). Three of these studies were prospective (Level-II) studies, whereas the remaining eleven were retrospective (Level-III) studies (Table II). A cumulative total of 706 patients were included in these studies; the mean duration of follow-up was 67.1 and 57.8 months for the operative and nonoperative treatment groups, respectively. A favorable clinical outcome (defined as good or better) was reported in 88% of the operatively managed patients compared with 85.5% of the nonoperatively managed patients. Anatomic reduction of the joint was achieved in 59% of the operatively managed patients and only 14.7% of the nonoperatively managed patients. Evaluating factors such as return to work and return to sport revealed a quicker recovery for the nonoperatively managed patients, who needed roughly half the time to return to the previous level of activity.
We did not find high-quality randomized controlled trials that compared nonoperative and operative treatment of type-III acromioclavicular joint dislocations. Furthermore, the available clinical studies that were evaluated in this critical analysis had different operative techniques for acromioclavicular joint reconstruction and variable indications for operative intervention, and there was limited use of validated outcome measures, all of which make the comparison among different clinical studies difficult and less meaningful. Our understanding regarding the indications for primary operative treatment of type-III acromioclavicular joint dislocation is limited. There is a subset of patients with type-III acromioclavicular joint dislocations who will benefit from primary surgery, but we do not have validated objective criteria to identify this subset. It was recently postulated that patients with therapy-resistant scapular dysfunction (persistent scapular dyskinesia after a trial of physical therapy) and an overriding clavicle on a cross-body adduction radiograph (an unstable type-III acromioclavicular joint dislocation) may benefit from early operative fixation3. However, future prospective studies are required to validate this hypothesis.
Timing of Operative Treatment
There is insufficient evidence to support either early or delayed operative treatment of complete acromioclavicular joint dislocations. The literature search yielded five retrospective comparative (Level-III) studies that have addressed this question20-24 (Table III). The cumulative analysis of these five studies revealed 135 patients with a mean age of thirty-five years undergoing early operative treatment and ninety patients with a mean age of 36.8 years undergoing delayed operative treatment. The mean duration of follow-up for the patients managed with the early and delayed procedures was 52.8 and 54.0 months, respectively. A favorable outcome, defined specifically in each investigation as a tool for subjective evaluation, was reported in 123 (91%) of the patients in the early treatment group compared with only sixty-six (73%) of those in the delayed treatment group.
On the basis of the aforementioned analysis and quality of studies, there is insufficient evidence to recommend the optimum timing of operative intervention for the treatment of acromioclavicular joint dislocations. The five studies included in this review involved several different methods of acromioclavicular joint reconstruction (e.g., pinning and the Weaver-Dunn procedure). This variation in the surgical technique for the acromioclavicular joint reconstruction can result in differences in outcomes and makes comparison among a small number of studies very difficult. Although these five retrospective studies (Level-III evidence) suggest that there may be a benefit of early repair as compared with delayed repair, the quality of the evidence substantially limits the strength and certainty of this conclusion.
Anatomic and Nonanatomic Reconstruction
Rationale Behind Anatomic and Nonanatomic Acromioclavicular Joint Reconstructions
The acromioclavicular and coracoclavicular ligaments act in concert to provide stability to the acromioclavicular joint complex. Cadaveric studies have demonstrated that selective sectioning of the acromioclavicular ligament (with the coracoclavicular ligaments intact) results in a substantial increase in anteroposterior translation but no substantial increase in superior translation25-29. Furthermore, the coracoclavicular ligaments see a substantial increase in force when the acromioclavicular ligament is transected25,29,30. The coracoclavicular ligaments provide primary restraint to superior translation29,30. In a cadaveric experiment, Mazzocca et al.30 demonstrated that the failure of coracoclavicular ligaments in response to a load follows a predictable pattern in an acromioclavicular ligament-deficient shoulder. In all of the specimens that were tested, the conoid ligament disrupted first, followed by the trapezoid ligament.
Historically, acromioclavicular joint reconstructions have focused on maintaining anatomic reduction of the acromioclavicular joint with use of acromioclavicular or coracoclavicular joint internal fixation (e.g., Kirschner wire, Steinmann pin, hook plate, Bosworth screw), distal clavicular resection, and/or nonanatomic ligament reconstruction. Recently, there has been increased interest in anatomic reconstruction of the coracoclavicular and acromioclavicular ligaments with use of allograft or autograft for the treatment of chronic acromioclavicular joint dislocations. Controlled laboratory studies involving cadaveric specimens have demonstrated that anatomic coracoclavicular ligament reconstruction provides superior biomechanical properties when compared with conventional nonanatomic ligament reconstruction (Weaver-Dunn procedure)31-36. Furthermore, acromioclavicular ligament reconstruction in combination with anatomic coracoclavicular ligament reconstruction provides superior rotational and translational stability to the acromioclavicular joint complex as recently demonstrated in the cadaveric studies by Michlitsch et al.36 and Beitzel et al.37.
Critical Analysis of Anatomic and Nonanatomic Reconstruction
We identified three studies that clinically evaluated an anatomic acromioclavicular joint reconstruction technique and compared it with a nonanatomic acromioclavicular joint reconstruction technique38-40 (Table IV). Tauber et al.38, in a Level-II study, prospectively evaluated twenty-four patients with acromioclavicular joint dislocations who were managed with either the Weaver-Dunn procedure (n = 12) or anatomic reconstruction of the coracoclavicular ligaments (n = 12). At a mean of thirty-seven months of follow-up, the patients with anatomic reconstruction demonstrated superior outcome scores (American Shoulder and Elbow Surgeons [ASES] and Constant scores) and improved resistance to vertical stress loading (p < 0.05). In a retrospective study, Fraschini et al.39 (Level III) compared anatomic and nonanatomic reconstruction of the acromioclavicular joint for the treatment of chronic acromioclavicular joint dislocation. The anatomic technique used artificial ligament for the acromioclavicular joint reconstruction, whereas nonanatomic reconstruction of the acromioclavicular joint was performed with a coracoclavicular cerclage using Dacron vascular graft. The acromioclavicular joint stabilization was performed in both groups with two Kirschner wires that were removed at three weeks. Both surgically managed groups demonstrated superior outcomes compared with the untreated control group, but nonanatomic reconstruction was associated with a higher complication rate. In the most recent study, by Eschler at al.40, an anatomic procedure in which a polydioxanone (PDS) sling was passed under the coracoid and through two clavicular drill-holes was compared with a nonanatomic open reduction and internal fixation procedure utilizing a hook plate. The investigators reported no significant difference with respect to the Constant scores in the hook-plate group (91.2 ± 2.2) compared with the PDS group (94.6 ± 1.0). Despite the lack of significant difference in functional outcomes, the authors reported a more accurate restoration of coracoclavicular distance in the hook-plate group as compared with the PDS group. The aforementioned studies provide insufficient evidence that anatomic treatment provides a greater benefit to the patient than a nonanatomic approach for the operative treatment of complete acromioclavicular joint dislocation.
Operative and Nonoperative Treatment of Type-IV, V, and VI Injuries
There is a general consensus that high-grade injuries (types IV, V, and VI) should be treated operatively; however, the available evidence is of poor quality to support this recommendation4,6,23,41-43. Similar to type-III acromioclavicular joint dislocations, there is no consensus regarding the gold standard operative procedure (anatomic or nonanatomic) or ideal timing of operative treatment (early or delayed)4,6,20,23.
Treatment Recommendations and Senior Author’s Preferred Treatment Algorithm
The recommendations for the treatment of acromioclavicular joint injuries, based on the critical analysis of the available literature, are presented in Table V. As per the senior author’s clinical algorithm, type-I and II injuries are treated nonoperatively. This treatment includes the use of ice and anti-inflammatory medications to reduce pain and inflammation associated with the injury in order to allow early shoulder range of motion and recovery of shoulder strength. We recommend minimal immobilization to our patients and encourage early use of the injured extremity.
The decision to treat type-III dislocations is guided by the activity level of the patient, the severity of symptoms, and physical examination findings. According to the senior author’s clinical algorithm, for non-athletes, nonoperative management is the first line of treatment, similar to the algorithm for type-I and II injuries, but the rehabilitation support is more aggressive, consisting of range-of-motion and then strengthening exercises of the shoulder and scapula. The majority of patients respond to this therapy, but if pain and/or shoulder dysfunction persist after a trial of physical therapy, operative intervention is considered. Multiple factors are considered when deciding between operative and nonoperative treatment of a type-III dislocation in an athlete. These factors include the type of sport (contact or overhead), the level of participation (recreation or competitive), the specific sport being played (baseball, basketball, football, hockey, volleyball, etc.), and the timing of treatment with respect to the sport season (off season or in season). All athletes with type-III injures are given a three-week trial of nonoperative management and, if they demonstrate improved pain control and increased use of the arm, they are allowed to return to sport, especially if they are in the competitive season. If the athlete demonstrates no or minimal improvement with a trial of nonoperative management, operative treatment is discussed with the patient. We recommend operative treatment to patients with high-grade dislocations (types IV, V, and VI)32,44-47.
Open anatomic coracoclavicular and acromioclavicular ligament reconstruction using allograft tissue is our surgical procedure of choice for the treatment of acute and chronic complete acromioclavicular joint dislocations. Postoperative immobilization is an important component of rehabilitation following anatomic coracoclavicular ligament reconstruction to protect the surgical reconstruction during the initial phases of healing. We routinely immobilize our patients in a platform brace (e.g., a DonJoy Lerman shoulder orthosis [DJO Global, Vista, California, or a Gunslinger shoulder orthosis [AliMed, Dedham, Massachusetts]) for six to eight weeks, irrespective of the surgical technique used. The postoperative rehabilitation starts with an initial phase to restore scapular control and shoulder range of motion, followed by a structured strengthening program along with a home exercise program. Return to contact sports is allowed after six months.
Conclusion and Future Directions
There is no gold standard technique for the operative treatment of complete acromioclavicular joint dislocation, but operative treatment is evolving. The anatomic reconstruction of the coracoclavicular and acromioclavicular ligaments has shown promise in cadaveric biomechanical studies and clinical studies. There are still unresolved issues, including the indications for the operative treatment of type-III acromioclavicular joint dislocation, the timing of the operative procedure (early or delayed), the type of graft for the reconstruction (autograft, allograft, or synthetic graft), the type of repair (anatomic or nonanatomic), and the type of procedure (arthroscopic or open). There is a need for future investigations with higher level of evidence and long-term follow-up to guide future treatment algorithms for acromioclavicular joint dislocation.
Source of Funding: No external source of funding was used for preparation of this manuscript.
Investigation performed at the Department of Orthopaedic Surgery, New England Musculoskeletal Institute, University of Connecticut School of Medicine, Farmington, Connecticut
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.
- Copyright © 2015 by The Journal of Bone and Joint Surgery, Incorporated