Volume 26, Issue 2 , Pages 149-160, February 2010
All-Arthroscopic Weaver-Dunn-Chuinard Procedure With Double-Button Fixation for Chronic Acromioclavicular Joint Dislocation
Article Outline
Purpose
We described a novel all-arthroscopic technique of coracoclavicular ligament reconstruction and reported the early clinical and radiologic results of this procedure.
Methods
We performed all-arthroscopic coracoclavicular ligament reconstruction in 10 consecutive patients (8 men and 2 women; mean age, 41 years) with a symptomatic chronic and complete acromioclavicular (AC) joint dislocation (Rockwood type III or IV). Four patients had undergone surgery previously: two had initial pinning of the acute AC joint separation, and two had a subsequent Mumford procedure. The surgical technique, performed entirely by arthroscopy, consisted of (1) rerouting the coracoacromial ligament with a bone block harvested from the tip of the acromion in a socket created in the distal clavicle (Chuinard's modification of the Weaver-Dunn procedure) and (2) augmenting the reconstruction with 2 titanium buttons connected by a heavy suture in a 4-strand configuration (Double-Button fixation; Smith & Nephew Endoscopy, Andover, MA). Patients were prospectively followed up for a mean of 12.8 months (range, 6 to 20 months).
Results
One patient had a superficial infection of the superior (clavicular) portal, which resolved with oral antibiotics. At the most recent review, all patients were satisfied or very satisfied with the cosmesis, and 9 of 10 returned to previous sports, including contact and overhead sports. All symptoms resolved (pain, shoulder weakness, paresthesia). The mean postoperative University of California, Los Angeles modified AC rating score was 16.5 points (range, 13 to 18 points) out of 20 points. The mean Subjective Shoulder Value improved from 36% (range, 0% to 70%) preoperatively to 82.5% (range, 70% to 100%) postoperatively (P = .005). The bone block was totally healed in the medullary canal in 8 cases and partially healed in 2. No loss of reduction was observed in any of the patients.
Conclusions
Our study shows that severe chronic symptomatic AC joint separations, defined as Rockwood types III through V, can be repaired entirely by arthroscopy safely and effectively by transferring the coracoacromial ligament with a bone block in the distal clavicle. The bone block transfer (Weaver-Dunn-Chuinard procedure) has the advantage of making the repair easier and stronger, and it provides bone-to-bone healing by use of free, autologous vascularized tissue. Double-Button fixation has the advantage of maintaining the reduction during the biological healing process. Although the durability of the reconstruction remains unproven, in our short-term follow-up we observed no loss of reduction and the functional and cosmetic results were uniformly good.
Level of Evidence
Level IV, therapeutic case series.
Acromioclavicular (AC) joint dislocation is common in athletes, especially in contact sports such as rugby, football, and hockey or after a fall while skiing or cycling. Depending on the magnitude of the injury, the coracoclavicular (CC) ligaments as well as the AC joint capsule and ligaments can be stretched, partially torn, or completely ruptured.1, 2, 3 The CC ligaments are responsible for the upward rotation of the clavicle and downward rotation of the scapula, during abduction and forward elevation, described by Codman4 as synchronous scapula clavicular rotation.
The management of acute dislocation of the AC joint remains controversial, and patients are treated conservatively, even in cases of severe displacement.5, 6 A recent survey of over 500 members of the American Orthopaedic Society for Sports Medicine indicated that more than 80% of respondents prefer nonoperative treatment as initial management.7 However, 20% to 40% of patients treated conservatively after an acute AC joint dislocation have unsatisfactory results, with residual pain during shoulder motion, paresthesia, loss of strength and fatigue with overhead activities, and/or cosmetic concerns.8, 9 When patients are seen more than 6 weeks after the initial injury, AC joint dislocation is considered to be chronic because there is either partial or total resorption of the CC ligaments. This makes direct ligament repair insufficient to stabilize the AC joint, and most authors recommend augmenting the repair with an allograft or autologous tissue.10
The most popular and widely used CC ligament reconstruction technique for chronic injuries was originally described by Weaver and Dunn11 in 1972. The Weaver-Dunn procedure involves detaching the coracoacromial (CA) ligament from the acromion and transposing it to the lateral end of the distal clavicle. The transferred ligament is then secured across the cancellous bone to the upper part of the clavicle. In their original series Weaver and Dunn reported a failure rate of 28%, and poor results have been reported in other series, with loss of reduction after surgery because of stretch or pullout of the transferred CA ligaments.5, 12 To reinforce the repair, Chuinard and colleagues13 proposed a procedure in 1986 in which an acromial bone block attached to the CA ligament is transferred into the medullary canal of the resected distal end of the clavicle.
For many years, the open Weaver-Dunn-Chuinard (WDC) procedure has been our procedure of choice for the treatment of patients with symptomatic chronic AC joint dislocation. However, the need for extensive anterior shoulder dissection has been a concern, because the procedure is often partially indicated for cosmetic reasons. This has led us to adapt the WDC procedure to be performed entirely by arthroscopy. Simultaneously, we have developed a system named the Double-Button (Smith & Nephew Endoscopy, Andover, MA) that is strong enough to maintain the CC reduction and protect the bone-ligament transfer during the time needed for biologic healing. This device consists of 2 titanium buttons, 1 over the clavicle and 1 under the coracoid, connected by a high-strength polyethylene suture in a 4-strand configuration. The purpose of this study is to describe this novel all-arthroscopic technique of CC ligament reconstruction and report the early clinical and radiologic results of the procedure. We hypothesized that this procedure would provide a safe and effective treatment option for chronic AC joint dislocation, with a stable AC joint reconstruction and good cosmesis.
Methods
Patients
We prospectively followed up all patients who underwent an all-arthroscopic WDC procedure for the treatment of a complete AC joint dislocation (Rockwood type III or VI) at our institution.14 Inclusion criteria for the study were (1) persistent symptoms for at least 3 months after primary operative or nonoperative treatment and (2) a minimum of 6 months' clinical and radiologic follow-up. Exclusion criteria included (1) concurrent shoulder pathology requiring operative treatment in the same sitting and (2) acute treatment of AC dislocations. In all cases the procedure was performed by the senior surgeon or under his control. The local institutional review board approved the study, and all patients gave written informed consent.
We identified 12 patients who met the inclusion criteria. Two patients were excluded because of coexisting shoulder pathology requiring concurrent surgery, with a Bankart repair in the first case and a rotator cuff repair and tenotomy of the long head of the biceps in the second. The remaining 10 patients (8 men and 2 women) had a mean age of 41 years (range, 19 to 52 years). The mean follow-up was 12.9 months (range, 6 to 20 months). The mean time from injury to definitive surgery was 19.8 months (range, 3 to 60 months). The most common reasons that our patients were seeking surgery were pain over the AC joint, a feeling of AC joint instability with popping or grinding in the joint, shoulder fatigue, paresthesia with dead-arm syndrome, and deformity of the shoulder girdle.
The Rockwood classification was applied based on clinical examination radiographs. Four patients had previous operative treatment: two had initial pinning of the acute AC joint separation, and two had a subsequent distal clavicle resection (Mumford procedure). It should be noted that none of the patients in this series was treated primarily at our institution; all were referred to us subsequently after failure of their primary treatment modality at another center. Arm dominance, Rockwood type, injury mechanism, mode of initial treatment, and prior shoulder surgery are shown in Table 1 for each patient.
Table 1. Patient Data
| Patient No. | Gender | Age (yr) | Injured Side | RW | Injury Mechanism | Primary Treatment | Time I-R (mo) |
|---|---|---|---|---|---|---|---|
| 1 | F | 49 | L | III | Scooter | Pins, 8wk | 11 |
| 2 | M | 47 | R⁎ | IV | Kitesurfing | Pins,3wk | 4 |
| 3 | M | 20 | R | IV | Scooter | Conservative | 33 |
| 4 | M | 36 | R⁎ | V | All-terrainvehicle | Conservative | 3 |
| 5 | M | 38 | R | III | Rugby(professional) | Mumford | 32 |
| 6 | M | 48 | L⁎ | IV | Skiing | Conservative | 11 |
| 7 | M | 42 | L | III | Motorcycling | Conservative | 9 |
| 8 | M | 48 | R⁎ | IV | Dirtbiking | Pins,6wk | 22 |
| 9 | F | 32 | L | V | Skiing | Mumford | 17 |
| 10 | M | 52 | L | IV | Caraccident | Conservative | 60 |
⁎Dominant side. |
Rationale for Procedure
In chronic AC dislocations the CC ligaments resorb and the AC joint becomes arthritic. Suture repair of the CC ligaments is therefore insufficient, and a ligament reconstruction is needed for their replacement. The CC ligaments have been shown to fail by avulsion or midsubstance tear at 589 N.15 The load to failure of the native AC joint complex has been reported to be 815 N, whereas it is only 483 N for the Weaver-Dunn procedure.16 This means that the Weaver-Dunn reconstruction alone is weak and may not be sufficient to resist the forces present at the shoulder. In fact, the graft ligament can be stretched or torn because of the weight of the arm alone.17, 18
These biomechanical considerations have led us to use a stronger construct for AC joint repair, first by reinforcing the transferred CA ligament with a piece of bone harvested from the acromion (Chuinard's modification of the Weaver-Dunn procedure) and second by protecting this bone-tendon graft with a solid CC fixation, mechanically superior to the native AC joint complex and CC ligaments.
We have designed a device for this purpose called the Double-Button, which allows maintenance of the CC reduction while the acromial bone fragment heals inside the distal clavicle. The Double-Button is made by connecting 2 EndoButtons (Smith & Nephew Endoscopy) together with a heavy nonabsorbable suture (No. 5 FiberWire suture [Arthrex, Naples, FL] or Ultrabraid [Smith & Nephew Endoscopy]) placed in a 4-strand configuration. The EndoButton is a titanium button (4 × 12 mm) used for femoral graft fixation in anterior cruciate ligament reconstruction in the knee and distal biceps repair in the elbow.19, 20 The strength of graft fixation provided by the EndoButton in anterior cruciate ligament reconstruction has been shown to be 1,086 N in single-cycle load-to-failure testing with low displacement under cyclic loading.19, 21 Made of high-strength polyethylene, No. 5 FiberWire sutures have been shown to have a tensile strength of 485 N.15, 22
In summary, the Double-Button device maintains the CC reduction with a tensile strength superior to the native CC ligaments and AC joint complex.15 In addition, failure to resect the arthritic distal clavicle in these cases may result in painful AC joint arthrosis.23, 24
The aims of the arthroscopic WDC procedure with Double-Button fixation are therefore (1) to resect the distal clavicle and create a socket in its medullary canal, (2) to reroute the CA ligament with a bone block (harvested from the tip of the acromion) inside the distal end of the clavicle, and (3) to protect the bone-ligament transfer during the time of bone healing by maintaining the reduction with the help of the Double-Button (Fig 1).
Figure 1.
Principles of all-arthroscopic WDC procedure with Double-Button fixation. (A) The CA ligament is harvested with a bone block from the tip of the acromion and rerouted in the distal end of the clavicle. (The arthritic distal clavicle has been resected and a socket drilled in its medullary canal.) (B) The bone-ligament transfer is protected for the time of bone healing by maintaining the reduction with the help of the Double-Button (2 titanium buttons connecting the clavicle and the coracoid with 4 strands of suture).
Operative Technique
Although we perform this procedure with the patient in the beach-chair position, it may be performed in the lateral decubitus position. The bony landmarks and portals of the shoulder are outlined with a sterile marker, including the distal clavicle, the coracoid, and the acromion. In addition to the standard posterior portal, 4 portals are used: an anterolateral portal for the arthroscope, a lateral portal to cut the tip of the acromion, an anteromedial (AM) portal for the inferior part of the guide, and a superior portal for the superior part of the guide (Fig 2).
Figure 2.
Five portals used for AC joint reconstruction (A) in a drawing and (B) marked on a shoulder: posterior (P), lateral (L), anterolateral (AL), AM, and superior (S).
The surgical technique is composed of 5 operative steps, all performed arthroscopically (Video 1, available at www.arthroscopyjournal.org).
Coracoid ExposureThe arthroscope is first inserted through a standard posterior portal for glenohumeral joint exploration. An AM portal is then created 1 cm distal to the tip of the coracoid. This portal is created with an outside-in technique by first placing a spinal needle in the rotator cuff interval, entering the joint at the superior margin of the subscapularis tendon. Visualization of the base of the coracoid requires resection of some of the rotator interval capsule. Once a window is created in the rotator interval, the arthroscope can be pushed into the anterior bursa. Resection of the soft tissue located at the base of the coracoid is performed with a VAPR radiofrequency device (DePuy Mitek, Raynham, MA) to avoid bleeding. The arthroscope is then transferred to the anterolateral portal, while the VAPR device is introduced into the AM portal to complete the exposure of the base of the coracoid. Usually, the upper border of the subscapularis tendon can be perfectly seen.
Acromionectomy and CA Ligament HarvestingThe tip of the acromion and the CA ligament are exposed. The VAPR device is used to perform an anterior bursectomy inferiorly and to separate the coracoid, acromion, and CA ligament from the deltoid muscle superiorly. The anterior acromion must be perfectly visualized after elevation of the deltoid and trapezius fascia. By use of a small, high-speed bur (StoneCutter; Smith & Nephew Endoscopy) introduced through the lateral portal, a distal osteotomy of the acromion is performed (Fig 3A). Great care must be taken to harvest only a small piece of acromion; this is best done by first placing 3 spinal needles on each side of the acromial tip (medial, lateral, and distal) to define its borders visually. The size of the rectangular bone block should be approximately 5 × 5 × 7 mm. Once the acromial osteotomy is done, 2 polydioxanone (PDS) sutures are placed through the CA ligament passing around the bone block. These sutures will be used first to mobilize and pull the bone fragment medially with the CA ligament and second to shuttle the definitive suture at the time of fixation.
Figure 3.
Resection of tip of acromion and cavitation of distal end of clavicle. (A) The osteotomy of the tip of the acromion is performed with a bur introduced through the lateral portal; (B) a socket is then drilled inside the distal clavicle to receive the CA ligament with the acromial bone block.
The VAPR device is then used to perform a subperiosteal elevation of the deltoid and trapezius muscles from the distal clavicle. The clavicle must be free of all soft tissue surrounding the superior surface over approximately the distal 3 cm. The distal 0.5 to 1 cm of the clavicle, which is usually arthritic, is resected with a motorized bur (StoneCutter; Smith & Nephew Endoscopy) introduced through the AM portal. The resection is performed in a slightly oblique manner, resecting more anteriorly than posteriorly and more inferiorly than superiorly to better accept the graft and avoid tenting of the transferred CA ligament over a sharp bony edge.
The medullary canal of the clavicle is then enlarged and deepened with the same high-speed bur introduced through the lateral portal. It is helpful to have an assistant push the clavicle down and forward so that the bur can pass under or in front of the acromion. The cancellous bone is removed from the distal end of the clavicle to a depth of 1.5 cm. The distal clavicular socket must be larger than the bone block, measuring approximately 10 × 7 × 15 mm (Fig 3B).
CC Reduction and FixationA 5-mm skin incision is made over the clavicle in line with the coracoid process (superior portal), approximately 3 cm medial to the AC joint. A pilot hole is made with an awl on the posterosuperior cortex of the clavicle at the level of the coracoid process, and a 2.4-mm drill-tipped guide pin is introduced in the pilot hole.
A specific guide, the CC Reduction Guide (Smith & Nephew Endoscopy), is placed through the AM portal onto the inferior surface of the base of the coracoid, as well as on the superior part of the clavicle after connection with the guide pin. The tip of the guide must be advanced medially until the probe-like tip of the guide falls off the medial border of the base of the coracoid. This allows centering of the drill hole in the strong and wide part of the coracoid base. The clavicle is then reduced and held by manipulation of the guide, and the guide pin is advanced carefully through the clavicle and the base of the coracoid. The tip of the guide pin is viewed as it exits the base of the coracoid.
A 4.5-mm cannulated drill is carefully advanced over the guide pin to create a tunnel through the clavicle and the base of the coracoid. When the tip of the drill bit is seen exiting bone, it is stopped and left in place. The guide pin is then removed, leaving the cannulated drill in place. A PDS suture is inserted proximally into the cannulation of the drill and is advanced until it exits the drill at the base of the coracoid. The PDS suture is retrieved distally with a grasper and pulled out of the AM portal. The 4.5-mm drill is removed while the PDS suture is left in place.
The PDS suture is used as a shuttle to drag the traction suture attached to the Double-Button from distal to proximal. By pulling on the traction suture, the proximal button is passed through the AM portal, first through the coracoid and then through the clavicle. It is advanced under direct visualization until the proximal EndoButton arrives above the clavicle. This button is then flipped and locked (Fig 4A). The distal button is then visualized beneath the coracoid (Fig 4B). While the surgeon maintains downward pressure on the clavicle, the sutures are tightened by pulling alternately on each strand and the sliding knot is cinched onto the distal EndoButton. By tightening the sutures, the pulley action of the Double-Button construct can be used to further reduce the CC distance. The surgeon can judge the quality of the reduction in the anteroposterior and superoinferior planes by direct arthroscopic visualization. The distance between the clavicle and the coracoid is usually small, around 5 mm. Although the 4-strand configuration of sutures is secured with a sliding knot, the degree of friction is sufficient to maintain the reduction without any risk of suture slippage. A series of alternating half-hitches are then placed with the help of a knot pusher through the AM portal to secure the construct. The knots are intentionally placed under the coracoid (and not over the clavicle) to avoid cosmetic and skin infection problems. This is why the Double-Button should be introduced inside the shoulder from bottom to top and not from top to bottom.
Figure 4.
Double-Button fixation. (A) The proximal titanium button is placed over the clavicle, and (B) the inferior button is placed under the base of the coracoid process.
A second hole is drilled through the superior cortex of the clavicle, 1 cm lateral to the superior button. The 2 strands of heavy nonabsorbable sutures used to advance and flip the Double-Button are retrieved inside the medullary canal of the clavicle. The sutures are used to bring the bone block inside the clavicular socket and suture the CA ligament (Fig 5). The skin is then closed with either sutures or strips.
Figure 5.
Fixation of bone-tendon graft inside socket of distal clavicle. (A) The CA ligament with the tip of the acromion is rerouted inside the distal end of the clavicle; (B) the same suture used for the Double-Button is rerouted inside the medullary canal through a hole drilled on the superior cortex of the clavicle; (C) the probe is on the superficial side of the rotator cuff, and the acromion (on the left) and the reduced clavicle (on the right) are perfectly aligned.
Postoperative Care
Postoperatively, the arm was placed in a sling for 3 to 4 weeks. During this time, patients were allowed to perform pendulum exercises and to use their arm for most activities of daily living, except driving a car and heavy lifting, which were prohibited during the first 2 months. Patients were permitted to remove their sling daily for bathing. Normal use of the elbow, wrist, and hand was encouraged to prevent elbow stiffness and excessive swelling of the upper extremity. Strengthening was started after 2 months, and return to contact or overhead sports was allowed between 3 and 6 months. Patients were routinely seen at follow-up visits by a rehabilitation medicine specialist at 6 and 12 weeks and the senior surgeon at 6 months postoperatively.
Outcome Assessment
All patients underwent a detailed clinical examination at a minimum of 6 months postoperatively by the senior author (Table 1). Plain radiographs were obtained at all follow-up visits. Three patients also had computed tomography scans at 6 months (Fig 6). The imaging studies were examined for consolidation of the bony transfer, signs of loss of reduction, or other complications such as fracture, hardware migration, or heterotopic ossification. Preoperative and postoperative active range of motion and Subjective Shoulder Value (SSV) were also recorded. The SSV is the patient's self-rated subjective assessment of his or her shoulder function as compared with a normal shoulder expressed as a percentage, and it has been shown to correlate well with the Constant score.25, 26 We also obtained a University of California, Los Angeles (UCLA) modified AC rating score at the most recent clinical follow-up.18 The UCLA score is expressed as a numeric value between 0 and 20. Because of the small sample size, the Wilcoxon test was used to compare the means of the range-of-motion values and preoperative and postoperative SSV scores.
Figure 6.
Three-dimensional computed tomography images showing perfect reduction of AC dislocation. The clavicle is perfectly aligned with the acromion in both (A) the vertical plane and (B) the horizontal plane. (C) A lateral view shows the reduction with one button over the clavicle and the other button under the coracoid process.
Results
All patients underwent an entirely arthroscopic procedure, with no conversion to open surgery. No intraoperative complications occurred. All patients were discharged after 2 days in the hospital.
There was no significant difference between preoperative and postoperative active elevation, external rotation, or internal rotation. However, there was an improvement in the mean SSV from 36% (range, 0% to 70%) preoperatively to 82.5% (range, 70% to 100%) postoperatively, which was statistically significant (P = .005). The mean UCLA modified AC rating score was 16.5 points (range, 13 to 18 points) out of 20 points (Table 2). All preoperative symptoms resolved (pain, shoulder weakness, or paresthesia). Cosmetically, a symmetric appearance of both shoulders without bony prominence was obtained in all patients (Fig 7). No recurrent AC instability occurred. No anterosuperior subluxation of the shoulder joint developed in any patient.
Table 2. Clinical Outcome Data
| Patient No. | SSV (%) | UCLA Modified AC Rating Score (Points) | Follow-up (mo) | Primary Sporting Activity | RTS | Comments | |
|---|---|---|---|---|---|---|---|
| Preoperatively | Postoperatively | ||||||
| 1 | 30 | 85 | 18 | 20 | Walking | Y | |
| 2 | 40 | 75 | 16 | 19 | Kite surfing, snowboarding | Y | |
| 3 | 0 | 85 | 18 | 16 | Tennis | Y | Superficial infection |
| 4 | 10 | 100 | 18 | 13 | Swimming, cycling | Y | Partial bony union |
| 5 | 55 | 70 | 15 | 15 | Professional rugby | N | |
| 6 | 70 | 80 | 15 | 13 | Skiing, tennis | Y | EndoButton migration |
| 7 | 35 | 80 | 18 | 11 | Swimming | Y | |
| 8 | 65 | 80 | 18 | 10 | Weight lifting, dirt biking | Y | Partial bony union |
| 9 | 30 | 90 | 16 | 6 | Fitness training | Y | |
| 10 | 25 | 80 | 13 | 6 | None | NA | |
| Mean | 36⁎ | 83⁎ | 16.5 | 12.9 | |||
| SD | 22.5 | 8.2 | 1.8 | 4.8 | |||
⁎P = .005. |
Figure 7.
Both shoulders showed good cosmesis at 6 months.
All patients except 1 returned to work by 3 months and to their preinjury sports at the most recent follow-up visit. The patient who did not return to work was a 38-year-old professional rugby forward who had a history of multiple shoulder injuries with radiographic and arthroscopic evidence of glenohumeral arthritis and who had undergone a Mumford procedure and tenodesis of the long head of the biceps 1 year before the WDC reconstruction. Despite being satisfied with the procedure, he remarked that his shoulder still felt too “fragile” to return to competitive rugby.
The bony transfer appeared only partially united in 2 patients, whereas all others appeared well consolidated. No loss of reduction or periarticular ossification was detected on radiography or computed tomography in any patient at last review, including the patients with partial union of the transfer and the patient with button migration.
Two complications were observed. One patient had a superficial infection of the superior portal, which resolved with oral antibiotics. One patient had lateral migration of the subcoracoid EndoButton, which occurred between the 3- and 6-month visits, possibly due to cutout of the sutures through the coracoid (Fig 8).
Figure 8.
Radiographs showing proximal migration of subcoracoid EndoButton. (A) The radiograph at 3 months shows the subcoracoid EndoButton in a slightly lateral position and an oblique trajectory toward the supraclavicular button, and (B) the radiograph at 6 months shows proximal migration of the EndoButton, probably due to cutout of the sutures through the lateral border of the coracoid.
Discussion
In addition to modifying the shoulder contour, severe chronic AC joint separations (defined as Rockwood types III through V) may become symptomatic (because of pain, shoulder weakness, or paresthesia), leading some patients to seek surgery. Although open surgery can be proposed to reduce AC separations and reconstruct the CC ligaments, patients are reluctant to accept such a surgery because of the shoulder scar that will replace the dislocated AC joint. Arthroscopic treatment of chronic AC joint separation therefore represents an attractive alternative to open surgery. This study confirms our hypothesis: severe, chronic AC joint separations can be reduced and repaired safely and effectively entirely by arthroscopy. The principal finding of our study is that this all-arthroscopic CC ligament reconstruction with bone block transfer of the CA ligament into the clavicle (Chuinard's modification of the Weaver-Dunn procedure), associated with Double-Button fixation, provides a stable AC joint reconstruction with good cosmesis at short-term follow-up.
In our technique definitive reconstruction of the CC ligaments is provided by the bone block transfer of the CA ligament into the distal end of the clavicle (Chuinard's modification of the Weaver-Dunn procedure). The bone block transfer has the advantage of making the repair easier and stronger, and it provides bone-to-bone healing and a nearly anatomic reconstruction by use of free, autologous vascularized tissue available onsite. In our modified arthroscopic technique, the AC reduction is maintained with 2 titanium buttons (EndoButton; Smith & Nephew Endoscopy), 1 placed over the clavicle and 1 placed under the coracoid, connected with a heavy, nonabsorbable suture in a 4-strand configuration. This fixation allows protection of the bone block transfer of the CA ligament while it heals in the distal end of the clavicle. The main advantage of this combined reconstruction, along with being minimally invasive, is that it provides a very strong but flexible fixation superior to the native CC ligaments and AC joint complex.15, 16 The construct is strong enough to support early range of motion and rehabilitation, as well as early return to work. Other advantages include no hardware removal, very low morbidity, excellent cosmesis, and the potential to be performed as an outpatient procedure.
We observed 2 complications in our series, both related to technical mistakes, but these had no effect on the final result. The first complication was a case of superficial skin infection of the superior (clavicular) portal that occurred early in the series, and it was related to the placement of bulky knots over the superior (clavicular) button. Other authors have already reported that such large knots irritate the skin and eventually erode through the skin with resulting local infection.27 Although this infection resolved with oral antibiotics and local wound care, this led us to later modify the surgical technique; we now drag the Double-Button from bottom to top (and not from top to bottom) so that the bulky knots are placed under the base of the coracoid (and not under the skin). Our arthroscopic observation has shown us that the subcoracoid space is large enough for the placement of the knots without interfering with the superior tendon of the subscapularis. The other complication was a progressive lateral migration of the subcoracoid EndoButton, observed on follow-up radiographs (Fig 7). Although the coracoid appeared intact on radiographs, this was probably due to cutout of the sutures through the lateral aspect of the coracoid. We believe that slight lateral malpositioning of both the coracoid and clavicle tunnels contributed to this by generating a laterally directed abrasive force against a narrow tunnel wall, which eventually gave in. We did not observe this phenomenon in any other patient, and we believe that it can be avoided by careful (centered) tunnel placement.
Few reports have proposed arthroscopic (or semi-open) solutions to treat AC joint dislocations.15, 28, 29, 30 Different materials have been used to maintain the CC reduction, including suture loops, sutures and anchors, and different types of metallic pins, screws, or buttons. Possible failures with recurrent clavicular subluxation or dislocation have been reported after some of these procedures, with failure rates up to 50%.5, 12, 28 Fixation failures have been found to be due to hardware breakage or migration, suture abrasion and breakage, or bone erosion because of the potential sawing action of the sutures through the clavicle or the coracoid. With our technique, the titanium buttons, placed over the clavicle and under the coracoid, are sufficiently large and provide strong maintenance of reduction with homogeneous distribution of load on the bone surface, which protects against the sawing effect of the sutures. In contrast, we have observed that the round proximal button of the Tight Rope device (Arthrex) is too small and can sink through the clavicle, leading to a loss of reduction. As already mentioned, the distal EndoButton must be centered under the base of the coracoid process. The 4-strand suture configuration is stronger than the native CC ligaments and has the enormous advantage of providing a pulley effect, which makes CC reduction easy and smooth. In addition, the CC reduction can be individualized according to each patient while the final reduction is maintained without slippage of sutures because of a stop-block effect. Additional knots are tied over the distal button securing the repair. Because the tunnel through the coracoid is placed at its very base, this reconstruction avoids the excessive anterior translation of the clavicle seen after procedures that use a suture loop passed around the coracoid. Finally, despite the strong fixation obtained, the rotational movement of the clavicle during abduction, elevation, and rotation is preserved.
Biomechanical studies have suggested that the CA ligament is a weak graft that has approximately 25% of the initial strength of the native CC ligaments and less than 50% of the appropriate stiffness.31, 32 One argument against the CA ligament transfer (i.e., against the Weaver-Dunn procedure) is that despite good control of superior displacement with cyclic loading, this graft has shown poor control of anterior and posterior translation.3 This has led some authors to recommend the use of a semitendinosus allograft, which—from a biomechanical standpoint—appears to approximate the native CC ligaments more closely.3, 16, 31, 32 Although the semitendinosus allograft is initially stronger compared with the modified Weaver-Dunn procedure, such a graft may present some potential drawbacks that surgeons should keep in mind: it is expensive, the tissue allograft is not composed of living tissue, there is a risk of disease transmission, and large holes must be drilled in the clavicle and the coracoid, increasing the risk of fracture. It must also be noted that none of the previously mentioned biomechanical tests has been performed with Chuinard's modification of the Weaver-Dunn procedure by use of a transfer of the CA ligament with an acromial bone block. In addition, the Double-Button provides stability in the vertical direction, whereas the CA ligament–bone transfer provides stability in the horizontal plane. As already mentioned, our procedure offers an initial stability that is stronger that the native CC ligaments, and later on, we believe that the autograft gradually remodels and becomes stiffer and stronger with increased forces (Wolff's law). Hopefully, long-term follow-up of our patients will allow us to confirm this hypothesis.
It has been suggested that the CA ligament can sometimes be too long and lead to a lax CC repair.13 This is why some technical details are important to consider, and we recommend the following: (1) harvest a small piece of acromion (5 × 5 × 7 mm) together with the CA ligament, (2) resect a minimal amount of distal clavicle (only 0.5 to 1 cm), (3) create a socket large and deep enough in the distal end of the clavicle (7 × 10 mm and approximately 2 cm deep), and (4) advance the bone block as far medial as possible in the medullary canal of the clavicle. With our technique, it is easy to drag the acromial bone block deep enough inside the clavicle because we use the suture that is attached to the superior button. For that purpose, the 2 strands of suture are shuttled inside the distal clavicular socket after a hole has been drilled on the superior cortex of the clavicle (1 cm lateral to the superior button).
Another criticism of the CA ligament transfer is that it may disturb the acromial arch. It is true that the CA ligament is an important secondary glenohumeral stabilizer preventing anterosuperior displacement of the humeral head in longstanding massive rotator cuff tears.33 Therefore transfer of the CA ligament into the clavicle is contraindicated in cases of massive, irreparable rotator cuff tear. However, chronic AC joint dislocations are usually seen in young or mature patients with good rotator cuff tendons, in whom harvesting the CA ligament is rarely a problem. We did not observe any problems related to shoulder function in our patients. We have observed that, once the clavicle is reduced, the distance between the rerouted CA ligament and the tip of the acromion is small (around 5 mm). We hypothesize that the evolution must be the same as after an acromioplasty: the CA ligament reattaches itself to the acromion, thus reconstructing the arch.
Obvious criticisms of our study include the small number of patients, the absence of a control group, and the short follow-up. Despite the short follow-up, we are encouraged by these results, because most failures of CC ligament reconstruction occur in the first 6 weeks after surgery.24 Medium- and long-term follow-up will show whether our results are durable, and we intend to publish these results when they are available. Now that we have shown that an all-arthroscopic CC reconstruction is feasible, safe, and—at least in the short term—effective, it may be possible to consider a formal prospective randomized study.
Conclusions
Our study shows that severe chronic symptomatic AC joint dislocations, defined as Rockwood types III through V, can be reconstructed safely by an all-arthroscopic CA ligament–bone transfer. Although the durability of the reconstruction remains unproven, in our short-term follow-up, we observed no loss of reduction and found uniformly good results. The arthroscopic WDC procedure with Double-Button fixation is currently our procedure of choice to repair severe chronic symptomatic AC joint dislocations. We will continue to study it and report on longer-term follow-up results as they become available.
Supplementary data
Video 1. Surgical technique.
Supplementary data.
References
- . Biomechanical study of the ligamentous system of the acromioclavicular joint. J Bone Joint Surg Am. 1986;68:434–440
- . The acromioclavicular capsule as a restraint to posterior translation of the clavicle: A biomechanical analysis. J Shoulder Elbow Surg. 1999;8:119–124
- . A biomechanical evaluation of an anatomical coracoclavicular ligament reconstruction. Am J Sports Med. 2006;34:236–246
- . The shoulder. In: New York: Krieger; 1934;p. 52–54
- . Conservative or surgical treatment of acromioclavicular dislocation (A prospective, controlled, randomized study). J Bone Joint Surg Am. 1986;68:552–555
- . Clinical and radiologic outcomes of surgical and conservative treatment of type III acromioclavicular joint injury. J Shoulder Elbow Surg. 2006;15:300–305
- . Type III acromioclavicular separation: Results of a recent survey on its management. Am J Orthop. 2007;36:89–93
- . Operation for acromioclavicular dislocation (A review of 29 cases treated by one method). J Bone Joint Surg Br. 1987;69:715–718
- . Complete acromioclavicular dislocations: Treatment with a Dacron ligament. Injury. 1991;22:173–176
- . Current treatment of acromioclavicular separations. Curr Opin Orthop. 2007;18:373–379
- . Treatment of acromioclavicular injuries, especially complete acromioclavicular separation. J Bone Joint Surg Am. 1972;54:1187–1194
- . Surgical treatment of complete acromioclavicular dislocations. Am J Sports Med. 1995;23:324–331
- . Bone block transfer of coracoacromial ligament in acromioclavicular injury. Clin Orthop Relat Res. 1986;272–277
- . Injuries to the acromioclavicular joint. In: Fractures in adults. Ed 4. Philadelphia: Saunders; 1996;p. 1341–1413
- . Arthroscopic anatomic stabilization of acromioclavicular joint dislocation. Oper Tech Sports Med. 2004;12:43–48
- . Anatomical acromioclavicular ligament reconstruction: A biomechanical comparison of reconstructive techniques of the acromioclavicular joint. Am J Sports Med. 2005;33:1723–1728
- . Surgical treatment of chronic acromioclavicular joint dislocation by modified Weaver-Dunn procedure. Knee Surg Sports Traumatol Arthrosc. 2001;9:307–312
- . Reconstruction of chronic and complete dislocations of the acromioclavicular joint. Clin Orthop Relat Res. 1998;347:138–149
- . Fixation of the graft in reconstruction of the anterior cruciate ligament. J Bone Joint Surg Br. 2005;87:593–603
- . A biomechanical comparison of EndoButton versus suture anchor repair of distal biceps tendon injuries. J Shoulder Elbow Surg. 2006;15:509–514
- . The fixation strength of six hamstring tendon graft fixation devices in anterior cruciate ligament reconstruction (Part II: Tibial site). Am J Sports Med. 2003;31:182–188
- . Sutures and suture anchors: Update 2003. Arthroscopy. 2003;19:985–990
- . Coracoid fracture associated with acromioclavicular dislocation. Clin Orthop Relat Res. 1975;108:165–167
- . Acromioclavicular joint reduction, repair and reconstruction using metallic buttons (Early results and complications). Tech Shoulder Elbow Surg. 2007;8:213–221
- . A clinical method for functional assessment of the shoulder. Clin Orthop Relat Res. 1987;214:160–164
- . Comparison of the subjective shoulder value and the Constant score. J Shoulder Elbow Surg. 2007;16:717–721
- . Arthroscopic reconstruction of the acromioclavicular joint using semitendinosus allograft: Technique and preliminary results. Tech Shoulder Elbow Surg. 2008;9:109–113
- . Arthroscopic reconstruction of the coracoclavicular ligaments for acromioclavicular joint separations. Oper Tech Sports Med. 2004;12:49–55
- . Arthroscopic treatment of acute and chronic acromioclavicular joint dislocation. Arthroscopy. 2005;21:1017.e1–1017.e8www.arthroscopyjournal.org
- . Techniques of arthroscopic Weaver-Dunn in chronic acromioclavicular joint dislocation. Tech Shoulder Elbow Surg. 2006;7:155–159
- . Biomechanical function of surgical procedures for acromioclavicular joint dislocation. Arthroscopy. 2004;20:237–245
- . A modified technique of arthroscopically assisted AC joint reconstruction and preliminary results. Clin Orthop Relat Res. 2008;466:639–645
- . Superior humeral dislocation (A complication following decompression and debridement for rotator cuff tears). Clin Orthop Relat Res. 1991;263:135–141
The authors report no conflict of interest.
Note: To access the video accompanying this report, visit the February issue of Arthroscopy at www.arthroscopyjournal.org.
PII: S0749-8063(09)00700-2
doi:10.1016/j.arthro.2009.08.008
© 2010 Arthroscopy Association of North America. Published by Elsevier Inc. All rights reserved.
Volume 26, Issue 2 , Pages 149-160, February 2010