ACJ Reconstruction / Stabilisation

The indication for ACJ reconstruction is a symptomatic acromioclavicular dislocation graded by the Rockwood classification. The decision to operate is driven entirely by the grade, by the demands of the patient, and by whether the injury is acute or chronic. Decision by grade - Grade I (AC ligament sprain, CC intact, no displacement): sling 1–2 weeks, early mobilisation, analgesia. Return to sport at 2–3 weeks. No surgery.

• Grade II (AC ligament torn, CC sprained, less than 25% superior displacement on stress views): sling 2–3 weeks, physiotherapy from week 2. Return to sport at 4–6 weeks. Symptomatic AC joint arthritis develops in up to 40% at 5 years and may require later distal clavicle excision.
• Grade III (both AC and CC torn, 25–100% superior displacement): genuinely controversial — see the table below.
• Grade IV (posterior displacement of the clavicle into the trapezius, with posterior skin tenting): requires reduction and stabilisation regardless of demand level.
• Grade V (100–300% superior displacement, with muscle tenting, severe deformity and possible brachial plexus traction symptoms): always operative.
• Grade VI (inferior, subcoracoid or subclavicular dislocation from a high-energy mechanism, often with rib fractures, pneumothorax or brachial plexus injury): rare, and always operative.

Imaging that drives the decision. The Zanca view is the gold-standard radiograph for ACJ grading: a 10–15° cephalic-tilt AP taken bilaterally on a single film at 50% reduced penetration, so the coracoclavicular (CC) distance can be compared side to side (normal 11–13mm bilaterally). Stress views holding 5–10kg weights may unmask displacement in equivocal Grade II/III cases. Significant CC injury is present when there is greater than 5mm of absolute asymmetry or a greater than 25% relative increase; Grade V is defined as a CC distance greater than double the contralateral side (100–300% displacement). Timing changes everything. The later the surgery, the more complex the reconstruction required:

Choosing the reconstruction. Once the decision to operate is made, the choice of construct is the next decision:

The goal is to restore the coracoclavicular relationship anatomically — reconstructing both the conoid and trapezoid ligament functions — while protecting the brachial plexus and subclavian vessels that lie immediately posteromedial to the coracoid, and finishing with a meticulous repair of the deltotrapezial fascia for horizontal-plane stability. The exposure is a superior saber-cut approach with a deltotrapezial split, laid out in full in the first steps below.

View larger

View larger

View larger

Reference technique — Weaver-Dunn (historic). Still examined as the original transfer: (1) saber incision and deltotrapezial split as above; (2) excise 1–1.5cm of distal clavicle; (3) identify the coracoacromial ligament from the acromion; (4) detach it from the acromion with a small bone block using an osteotome, preserving the coracoid origin; (5) gouge a canal 1–1.5cm deep in the distal medullary clavicle; (6) pass the CA ligament into the canal and secure it with multiple number 2 non-absorbable sutures through drill holes in the clavicle; (7) augment with a CC suture around the coracoid and through the clavicle (the modified Weaver-Dunn); (8) repair the deltotrapezial fascia. It is biomechanically weak — the CA ligament transfer restores only about a quarter of the native CC load-to-failure — which is its high loss-of-reduction rate. Arthroscopic-assisted technique (contemporary evolution). Several centres now perform the reconstruction arthroscopically-assisted or fully arthroscopic: the coracoid tunnel is drilled under arthroscopic visualisation from a posterior glenohumeral portal, giving direct confirmation of tunnel position and lowering coracoid fracture risk, while the clavicle tunnels are still made through a small superior open incision with fluoroscopy. Advantages are direct visualisation of the coracoid base, concurrent treatment of intra-articular pathology (SLAP tears, rotator cuff disease), and smaller incisions with equivalent reduction and functional outcomes (Salzmann 2010, PMID 20442326). The trade-offs are a steep learning curve, longer operative time during learning, and the continued need for a superior clavicle mini-incision. After securing any construct, perform dynamic fluoroscopy — a Zanca view through abduction and cross-body adduction — to confirm reduction is maintained through the functional arc.

Rehabilitation principle — progressive loading without traction. For the first 6 weeks the construct is protected from the deforming force of arm weight: gravity pulls the arm (and clavicle) inferiorly relative to the scapula, so the sling supports the forearm and elbow to offload the reconstruction. | Phase | Timing | Immobilisation | Therapy | |-------|--------|----------------|---------| | 1 — Protection | 0–6 weeks | Broad arm sling worn for 4–6 weeks | Elbow/wrist/hand exercises from day 1; pendulums from week 2. No active abduction, no overhead motion, no lifting greater than 1kg, no cross-body adduction | | 2 — Mobilisation | 6–12 weeks | Sling off at 6 weeks | Active-assisted forward flexion and abduction in a pain-free range; periscapular (serratus, lower trapezius) activation; rotator cuff resistance band from week 8 | | 3 — Strengthening | 12–24 weeks | — | Full active ROM expected by week 12; progressive resistance; overhead press and bench press deferred to week 16; sport-specific training from week 16 | Hamstring donor site (if applicable): knee ROM from day 1, quadriceps strengthening from week 2, no running until week 8. | Activity | Timeframe | |----------|-----------| | Driving (manual) | 6–8 weeks | | Office work (one-handed) | 2–4 weeks | | Light manual work | 12 weeks | | Heavy manual work | 4–6 months | | Contact sport | 4–6 months | | Overhead sport | 5–6 months | | Full unrestricted return | 6 months | Physiotherapy principles. In the protection phase, therapy focuses on cervical and periscapular re-education (posture correction and lower trapezius activation to avoid shoulder protraction, which increases CC distraction forces), rotator cuff activation at zero load (external rotation isometrics in the sling from week 2), and proprioception (joint position sense is disrupted after high-grade injuries; gentle joint mobilisation from week 4 aids neuromuscular re-education). Expected outcomes. Rockwood III (operative group): 85–90% good-to-excellent results at 2 years, mean ASES improvement of 25–30 points, CC distance restored to within 3–5mm of the contralateral side in 75–80%, and return to pre-injury activity in 80–85%. Rockwood V: 80–85% satisfactory outcomes, with any pre-operative neurovascular symptoms resolving in 85–90% after reduction and a higher risk of secondary AC arthritis than Grade III. Chronic reconstruction (greater than 3 months): anatomic CC reconstruction with a free tendon graft gives good-to-excellent outcomes in those who avoid a complication (mean ASES around 94 at 2 years; Millett 2015, PMID 25998014), but the procedure is technically more demanding and carries a clinically meaningful revision-requiring complication rate (around 22%); roughly 75–80% return to pre-injury function. Complications

The Rockwood classification in full. Rockwood graded ACJ injuries by the structures torn and the direction and degree of displacement, and the grade drives management:

Coracoclavicular ligament anatomy (the critical stabiliser). The CC ligament is the primary restraint to vertical displacement and comprises two distinct parts. The conoid ligament is posteromedial and cone-shaped, with a broad base on the posteromedial coracoid and its apex on the conoid tubercle of the clavicle about 45–55mm from the AC joint; its vertical collagen fibres make it the primary restraint to superior translation. The trapezoid ligament is anterolateral and trapezoidal, arising from the superior anterior coracoid and inserting on the trapezoid line of the clavicle about 25–35mm from the AC joint, anterior to the conoid insertion; its horizontal fibres make it the primary restraint to axial compression and lateral displacement. The 3–5mm interligamentous space between them contains a fat pad and small communicating vessels and is the anatomic plane for graft passage. The coracoacromial ligament runs from the anterolateral coracoid tip to the undersurface of the acromion, forming the coracoacromial arch; it has no role in AC joint stability (it is a roof, not a tether) and is harvested for the Weaver-Dunn transfer. Bony anatomy. The AC joint is a diarthrodial joint between the lateral clavicle (convex) and the medial acromial facet (flat to concave), with an average joint gap of 3–5mm and an intra-articular disc present in about 70% that degenerates with age and is gone by the fourth decade. Acromial morphology is described by Bigliani as flat (Type I), curved (Type II) or hooked (Type III, associated with impingement). The distal clavicle is elliptical in cross-section and normally overrides the acromion by 0–3mm in the coronal plane. Deltotrapezial fascia. The deltotrapezial fascia (deltotrapezial aponeurosis) is the conjoined fascia of the deltoid origin (anteroinferior) and the trapezius insertion (posterosuperior) at the clavicle and acromion. It is the primary incision layer in a superior exposure, a structural contributor to horizontal-plane stability, and the layer that must be repaired at closure to restore the dynamic muscular stabilisers. Relevant neurovascular anatomy. The suprascapular nerve (upper trunk, C5–C6) passes through the suprascapular notch to supply supraspinatus and infraspinatus; it is not directly at risk unless dissection extends far medially. The lateral pectoral nerve (lateral cord, C5–C7) runs with the pectoral branch of the thoracoacromial artery and is at risk with excessive medial dissection around the coracoid base. The musculocutaneous nerve enters the conjoint tendon 3–8cm distal to the coracoid tip. The subclavian artery and vein lie posterior to the clavicle in the costoclavicular space and are relevant during clavicle tunnel drilling. Chronic superior clavicle displacement in Rockwood V can cause thoracic outlet syndrome from brachial plexus traction, which resolves after reduction. Biomechanics of ACJ stability. Vertical (superior-inferior) stability is provided primarily by the conoid ligament (force to failure about 400–600 N), with the trapezoid a secondary vertical restraint and the deltotrapezial fascia a minor supplementary check. Horizontal (anteroposterior) stability is provided primarily by the AC ligaments, particularly the superior and posterior bands, which resist AP translation and axial rotation; the CC ligaments provide minimal horizontal restraint. This distinction is critical — reconstruction must address BOTH planes. Rotational stability is resisted by the posterior AC capsule and posterior deltotrapezial fascia, which is why the double-tunnel (conoid + trapezoid position) technique is biomechanically superior to a single-tunnel construct. | Structure | Approximate load to failure | |-----------|-----------------| | Conoid ligament (isolated) | ~500 N | | Trapezoid ligament (isolated) | ~300 N | | Intact CC ligament complex | ~500–725 N | | Coracoacromial ligament transfer (Weaver-Dunn) | ~100–150 N (about 25% of native CC) | | Single suspensory cortical button construct | ~500–1000 N (initial, construct-dependent) | | Doubled semitendinosus / looped tendon graft | Exceeds the native CC complex on cyclic testing | Figures are approximate; cadaveric values vary by specimen age, loading direction and construct (see Mazzocca 2005, PMID 16282577). Clinical implication. The Weaver-Dunn coracoacromial ligament transfer restores only a fraction of native CC strength, which explains its high loss-of-reduction rate. Modern suspensory and biological graft techniques more closely approximate or exceed native biomechanical behaviour, and on cyclic loading the anatomic free-graft reconstruction restores anteroposterior translation closer to the intact state than the Weaver-Dunn (Mazzocca 2005, PMID 16282577).

1. Weaver JK, Dunn HK. "Treatment of acromioclavicular injuries, especially complete acromioclavicular separation." J Bone Joint Surg Am. 1972;54(6):1187-1194. PMID: 4652050. [Original description of the coracoacromial ligament transfer to the resected distal clavicle] 2. Mazzocca AD, Arciero RA, Bicos J. "Evaluation and treatment of acromioclavicular joint injuries." Am J Sports Med. 2007;35(2):316-329. PMID: 17251175. doi:10.1177/0363546506298022. [Comprehensive review of ACJ anatomy, biomechanics, classification and treatment; emphasises horizontal as well as vertical instability and the limitations of the Weaver-Dunn] 3. 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(2):236-246. PMID: 16282577. doi:10.1177/0363546505281795. [Cadaveric study: anatomic free-graft CC reconstruction has less anterior/posterior translation and more closely approximates the intact joint than the modified Weaver-Dunn] 4. 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(5):811-817. PMID: 17293463. doi:10.1177/0363546506297536. [Defines the conoid and trapezoid footprints: conoid about 47mm (M) / 43mm (F) and trapezoid about 25mm (M) / 23mm (F) from the lateral clavicle; conoid/trapezoid-to-clavicle-length ratios constant at about 0.31 and 0.17] 5. Beitzel K, Cote MP, Apostolakos J, et al. "Current concepts in the treatment of acromioclavicular joint dislocations." Arthroscopy. 2013;29(2):387-397. PMID: 23369483. doi:10.1016/j.arthro.2012.11.023. [Systematic review: consensus for non-operative type I-II and initial non-operative type III, operative type IV-VI; insufficient evidence for early vs delayed and anatomic vs non-anatomic] 6. Millett PJ, Horan MP, Warth RJ. "Two-year outcomes after primary anatomic coracoclavicular ligament reconstruction." Arthroscopy. 2015;31(10):1962-1973. PMID: 25998014. doi:10.1016/j.arthro.2015.03.034. [31 shoulders (Rockwood III and V) with free tendon allograft; ASES improved 58.9 to 93.8; 22.6% required revision surgery for graft attenuation, clavicle fracture, distal clavicle hypertrophy or adhesive capsulitis] 7. Martetschläger F, Horan MP, Warth RJ, Millett PJ. "Complications after anatomic fixation and reconstruction of the coracoclavicular ligaments." Am J Sports Med. 2013;41(12):2896-2903. PMID: 24007761. doi:10.1177/0363546513502459. [59 procedures; overall complication rate 27.1%; coracoid fracture, clavicle fracture and graft rupture described; construct survivorship 86.2% at 12 months and 83.2% at 24 months] 8. Ceccarelli E, Bondi R, Alviti F, Garofalo R, Miulli F, Padua R. "Treatment of acute grade III acromioclavicular dislocation: a lack of evidence." J Orthop Traumatol. 2008;9(2):105-108. PMID: 19384625. doi:10.1007/s10195-008-0013-7. [Systematic review: clinical results comparable between operative and non-operative grade III, with more complications in the surgical group; non-operative treatment remains valid first-line] 9. Tamaoki MJS, Lenza M, Matsunaga FT, Belloti JC, Matsumoto MH, Faloppa F. "Surgical versus conservative interventions for treating acromioclavicular dislocation of the shoulder in adults." Cochrane Database Syst Rev. 2019;10:CD007429. PMID: 31604007. doi:10.1002/14651858.CD007429.pub3. [Cochrane review, 6 trials, 357 patients: no difference in 1-year DASH function; conservatively treated patients recovered faster at 6 weeks; higher adverse-event rate with surgery (RR 2.82)] 10. Salzmann GM, Walz L, Buchmann S, Glabgly P, Venjakob A, Imhoff AB. "Arthroscopically assisted 2-bundle anatomical reduction of acute acromioclavicular joint separations." Am J Sports Med. 2010;38(6):1179-1187. PMID: 20442326. doi:10.1177/0363546509355645. [23 patients (mostly Rockwood V); Constant score improved 34.3 to 94.3 at 24 months; supports the anatomic 2-tunnel/double-bundle concept and stresses precise tunnel and button placement]