Acromioclavicular Joint Injuries

Injuries to the acromioclavicular joint account for 9-12% of shoulder injuries and occur particularly in young active males in contact sports, with a male to female ratio of about 5:1. They result from disruption of the ligaments that stabilise the joint.

Mechanism. The common injury is a direct blow: a fall onto the point of the shoulder with the arm adducted. Less often the force arrives indirectly, through a fall onto the outstretched hand with the force transmitted to the shoulder. The settings are contact sports such as rugby, AFL and ice hockey, and cycling crashes.

Risk factors.

• Contact sports participation
• Male gender
• High-velocity activities (cycling, motorcycling)
• Previous AC joint injury

Scope note. This page covers the general classification, biomechanics and management of AC joint injuries. For sport-specific decision-making, in-season management and return-to-play criteria in athletes, see AC joint injuries in athletes.

The joint. A synovial diarthrodial joint between the lateral clavicle and the medial acromion, containing a fibrocartilaginous intra-articular disc that degenerates with age. The capsule is reinforced by the AC ligaments. The contact area is small, so the stress across it is high.

Two planes, two sets of ligaments. The AC ligaments give the joint its horizontal (anteroposterior) stability; the coracoclavicular ligaments give its vertical stability. This division is what the Rockwood grades measure: an AC-ligament injury with intact CC ligaments is low-grade, and once the CC ligaments are disrupted the clavicle is free to displace.

View larger

The AC ligaments. There are four, superior, anterior, posterior and inferior, reinforcing the capsule. The superior ligament is the one that matters most: it blends with the deltotrapezial fascia to form a strong superior stabiliser, and the posterior ligament is the next most important. The percentages that get quoted for each come from two experiments that asked different questions, which is why they appear to disagree.

The CC ligaments. The trapezoid is lateral, quadrilateral and the weaker of the two; the conoid is medial, conical and the stronger. Both run from the base of the coracoid process to the undersurface of the clavicle, and together they are the vertical restraint that is lost in Type III and above. Because they act together, an anatomic reconstruction has to recreate both.

The ratio is what transfers. Rios (PMID 17293463; 120 dry and 19 fresh cadaveric clavicles) found the conoid at 0.24-0.31 and the trapezoid at 0.17 of total clavicle length, constant across sex and race. Clavicle length varies, so measure the clavicle intraoperatively and the tunnel positions follow.

View larger

The muscles. The deltoid originates from the lateral clavicle and the acromion anteriorly; the trapezius inserts on the clavicle and acromion posteriorly. Between them the deltotrapezial fascia is critical both to stability and to surgical repair.

Tossy (1963) described Types I-III. Rockwood expanded the scheme to Types IV-VI, recognising displacement patterns that require surgical intervention. Each grade records which ligaments have failed and where the clavicle has gone, and the treatment follows from that.

View larger

History. The mechanism is usually a fall onto the point of the shoulder or a direct blow. Beyond that, the answers that change management:

• Time since injury
• Sport and occupation
• Hand dominance
• Overhead requirements (throwing athlete, labourer)
• Previous AC joint problems

Examination. Tenderness over the joint and pain on cross-body adduction are present at every grade and tell you only that the AC joint is the source. The findings that grade the injury are the step-off, the piano key sign and horizontal instability.

Cross-body adduction (scarf) test. The patient reaches the hand to the opposite shoulder; pain at the AC joint is positive. It is sensitive but not specific for AC injury.

O'Brien (active compression) test. The arm is held at 90° flexion and 10° adduction, internally rotated with the thumb down, and the patient resists a downward force. Pain at the AC joint that is relieved with supination is positive. The test is also positive in SLAP lesions.

Piano key sign. The examiner presses the lateral clavicle inferiorly; it depresses and then springs back. The sign tests vertical stability, and a frank piano key, marked depression with rebound, indicates CC ligament compromise, meaning Type III or above. In a Type II a minor piano key may be present, because the torn AC ligaments let the clavicle ride slightly high and be depressed, but the sprained-yet-intact CC ligaments limit true vertical translation.

Horizontal instability. Stabilise the acromion and translate the clavicle anteroposteriorly; increased translation means the AC ligaments are torn. This is a separate test from the piano key, which says nothing about the horizontal plane.

Associated injuries. Look for a scapula fracture (floating shoulder), a clavicle shaft fracture and rotator cuff pathology, examine neurovascular status, and check for skin tenting, which is a relative surgical indication. High-grade dislocations also carry an appreciable rate of occult intra-articular pathology: arthroscopy in surgically treated Type III-V injuries found intra-articular lesions in 18.2%, including SLAP lesions in 14.3%.

Differentials. For the painful or deformed superolateral shoulder, each entry has a discriminating feature and a test.

AP and Zanca views. The Zanca view is an AP with 10-15° of cephalic tilt centred on the AC joint. The tilt angles the beam perpendicular to the plane of the joint and takes the spine and scapula out from behind it, which is what makes it the optimal view for the joint. On it, compare the CC distance with the other side, measure the AC joint width and assess the percentage displacement.

View larger

Axillary lateral view. It shows the hook of the acromion, the coracoid and the glenoid, and it is the only view on which posterior displacement can be seen: in a Type IV the clavicle lies posterior to the acromion.

Stress views. Weighted views, with 5-10kg in the hands, may help to differentiate a Type II from a Type III and to quantify displacement. Most centres no longer use them routinely, and pain limits their usefulness in the acute setting.

View larger

Measurements. The numbers that grade the film:

CT is for the bone: a suspected fracture of the clavicle, acromion or coracoid, a failure of conservative treatment in which an occult fracture is being sought, and planning before a complex reconstruction.

MRI is for a suspected associated rotator cuff injury, chronic AC joint pain where the disc and any arthritis need assessing, and preoperative assessment before a delayed reconstruction.

The decision. Types I and II are treated conservatively and Types IV to VI surgically. Type III is the argument.

Type III. The current consensus is conservative first: 80-90% of Type III injuries treated conservatively achieve a satisfactory outcome, and multiple RCTs show no difference between operative and non-operative treatment. Pooling the four RCTs restricted to Type III (244 patients) found no significant long-term difference in Constant score (MD 4.82, 95% CI -6.42 to 16.06); surgery gave better early pain relief and a better coracoclavicular distance, at the cost of more complications. Surgery is considered in throwing athletes, heavy labourers and overhead workers, and in the patient with persistent symptoms at 3-6 months, which is uncommon, and only after discussing the risks and benefits. Do not reflexively operate.

Stable or unstable: IIIA and IIIB. The ISAKOS upper-extremity consensus (2014) subdivides Type III by dynamic stability and shoulder function rather than treating it as one group:

• Type IIIA, stable: no overhead or throwing dysfunction, no scapular dyskinesis, and the deformity reduces or is non-disabling; a therapy trial confirms good function. Conservative management. This is the larger group, which is why the trials show no difference: most Type IIIs are functionally IIIA.
• Type IIIB, unstable: persistent overhead dysfunction, scapular dyskinesis or winging, and therapy-resistant symptoms, often with horizontal instability. Consider surgical stabilisation or reconstruction.

The practical sequence is to treat essentially every Type III conservatively for a period with a structured physiotherapy and scapular programme, then re-classify: those who recover function are IIIA and continue non-operatively, those who remain symptomatic with dyskinesis are IIIB and are offered surgery. That is how the literature can show no overall difference while selected patients clearly benefit. In the exam, do not say "Type III is controversial" and stop; say you would trial rehabilitation and subdivide.

After conservative treatment. Persistent pain affects 10-20% and is usually mild, and AC joint arthritis is the long-term risk. The other failure is the patient who needs a delayed reconstruction, and the randomised data put this higher than the usual 5-10% quote: in the ACORN trial of 60 patients aged 16-35 with acute Type III/IV injuries, 5 of the 30 non-operative patients failed and needed delayed reconstruction (Level I). Surgery in that trial produced a far better radiographic reduction (mean residual displacement 1.75mm versus 10.61mm) yet no functional advantage at 1 year (DASH 5.63 versus 4.67), at over four times the cost. The honest counselling position is that conservative care is the right first line, the deformity will usually persist, and a minority will still come to surgery later.

After a hook plate. The complications are the hook's: impingement, acromial erosion and fracture, cuff damage, the mandatory removal, and recurrent instability in 10-15% once the plate is out. The rates are in Surgical Technique.

After CC reconstruction.

• Loss of reduction or failure (10-15%)
• Coracoid fracture during drilling (2-5%)
• Clavicle fracture through the tunnels
• Graft-site morbidity with autograft
• Infection (1-2%)
• Nerve injury (rare: musculocutaneous, suprascapular)
• Heterotopic ossification, which is common

After CC reconstruction the programme runs in four stages:

After a hook plate the early protocol is the same, with hardware removal at 3-4 months built into it. After removal the shoulder is protected for 2-4 weeks and then activity progresses. Some loss of reduction after removal is normal and usually asymptomatic.

Conservative treatment. Types I and II do excellently, with full recovery expected, and most patients return to full sport and activity.

Surgery. Types IV-VI do well with surgery, with 90% satisfaction. A Type III operated on late, after failed conservative treatment, does as well as one operated on early.

The bump. A persistent cosmetic deformity is common after conservative and after surgical treatment. It is usually well tolerated and does not correlate with functional outcome, so counsel patients about it preoperatively.

What decides the outcome.

• Accuracy of diagnosis (do not miss a Type IV)
• Patient selection between conservative and surgical treatment
• Surgical technique (anatomic reconstruction preferred)
• Rehabilitation compliance
• Associated injuries (rotator cuff, fractures)

Global epidemiology:

• In a population-based Scottish study, the acromioclavicular joint accounted for an incidence of 8.9 dislocations per 100,000 per year (Hindle et al., Injury 2013), making it one of the more frequently dislocated joints of the appendicular skeleton.
• Strongly male-predominant and concentrated in the young active population (mean age mid-30s in operative series), with a peak in the second-to-fourth decades.
• Mechanism is typically a direct blow to the point of the shoulder (contact and collision sports - rugby codes, American football, ice hockey, AFL) or a cycling/motorcycling fall; the same mechanism predominates internationally.
• Most injuries are low-grade (Rockwood I-II); complete (III-VI) dislocations are the minority but generate most of the surgical workload and controversy.

Side-by-side guideline and society positions: