Proximal Humerus Fracture - Plating vs Hemiarthroplasty

Indication. Surgery is considered for a proximal humerus fracture that is sufficiently displaced to predict a poor result in a sling, in a patient whose bone quality and demands justify fixation. The decision rests on the Neer pattern, bone quality, patient age and demand, and head viability — not on displacement alone. A fragment only counts as a separate "part" when it is displaced more than 1 cm or angulated more than 45 degrees. Most minimally displaced and 1-part fractures (about 80 percent) are managed non-operatively; the surgical decision is really about 2-, 3- and selected 4-part patterns. Image before you decide. AP, scapular Y and axillary radiographs are the minimum; for any operative case obtain a CT with 3D reconstruction to read the fracture personality — is there a head-split component, how big and how displaced are the tuberosities, what is the bone quality, and critically is the medial hinge intact (valgus-impacted) or disrupted (displaced). That single distinction drives both the AVN risk and the choice between plating and arthroplasty (the full classification is set out in Background & Evidence). The four treatment tracks. Whatever is chosen, the exposure is the same deltopectoral approach; the divergence is what you do once you are there:

The evidence behind the decision. The PROFHER trial (Rangan, JAMA 2015) randomised 250 adults with displaced surgical-neck fractures to surgery (fixation or humeral head replacement) versus a sling and found no difference in Oxford Shoulder Score over two years (39.07 versus 38.32; difference 0.75, 95 percent CI -1.33 to 2.84), with no difference in complications or secondary surgery — so routine surgery is not superior to non-operative care for most displaced surgical-neck fractures. The Cochrane review (Handoll, 2015) found insufficient evidence for surgical superiority and a higher reoperation burden with surgery. When arthroplasty is indicated in the elderly, the Sebastia-Forcada RCT (JSES 2014) showed reverse shoulder arthroplasty superior to hemiarthroplasty in patients over 70 (Constant 56.1 versus 40.0; forward elevation 120 degrees versus 80 degrees), with function independent of tuberosity healing — the basis for reverse increasingly replacing hemiarthroplasty. Named-society guidance (AAOS Appropriate Use Criteria, BOA/BOAST) supports non-operative management as the default for most surgical-neck fractures, reserving arthroplasty for displaced 4-part and head-split patterns in lower-demand elderly patients. Consent specifically for the high overall complication rate (30-50 percent): AVN (15-75 percent by pattern), tuberosity nonunion or pulloff (10-20 percent after ORIF, 15-30 percent after hemiarthroplasty), stiffness and adhesive capsulitis (30-40 percent), varus collapse (10-20 percent), screw penetration (5-15 percent), axillary nerve palsy and infection (1-2 percent), and the possibility of conversion to arthroplasty. Setup. Beach-chair position, arm draped free and mobile, radiolucent attachment for intra-operative fluoroscopy. General anaesthesia with an interscalene block gives excellent post-operative analgesia and allows day-1 therapy. Confirm fluoroscopy access for AP, scapular Y and axillary views before prepping.

The goal is to restore the head-shaft-tuberosity relationship and stable fixation through the deltopectoral approach while protecting the tenuous head blood supply and the axillary nerve. The exposure is laid out in full below as the first steps of the sequence (and in depth on the deltopectoral approach to the shoulder page). When the decision is arthroplasty, the same exposure and fracture exposure are shared and the sequence diverges at head removal.

View larger

Rehabilitation The balance is early passive motion (to prevent stiffness) against protecting tuberosity fixation (to prevent pulloff). An interscalene block makes day-1 therapy possible. | Phase | Timing | Immobilisation | Therapy | |-------|--------|----------------|---------| | Immediate | 0-2 weeks | Sling with abduction pillow (20-30 degrees) | Pendulums from day 1; interscalene catheter 48-72 h; wound check day 1-2; radiographs at 2 weeks | | 1 | 2-6 weeks | Sling between exercises, weaning after 4 weeks | Passive ROM only — flexion, external rotation, internal rotation to tolerance; no active motion | | 2 | 6-12 weeks | Wean sling | If tuberosities are healing on the 6-week film: active-assisted ROM (pulley, wand, table slides), gentle isometrics | | 3 | 12+ weeks | — | Active ROM and progressive resistance (Theraband, light weights); rotator-cuff strengthening | Radiographs at 2 weeks (wound and reduction), 6 weeks (tuberosity healing), 12 weeks (union), 6 months and 1 year (AVN surveillance). Motion recovery is slow and may continue for 12-18 months; most return to desk work by 6 weeks, daily activities by 4-6 months and sport by 6-12 months. Complications The overall complication rate is high (30-50 percent) even with good technique.

Epidemiology. Proximal humerus fractures are common in older adults, with a strong female predominance and a peak in the seventh decade and beyond, typically after a low-energy fall onto the outstretched hand; about 80 percent are minimally displaced (Neer 1-part) and are treated non-operatively. Why the medial hinge matters (blood supply and AVN). The humeral head receives blood principally from the arcuate artery, the ascending branch of the anterior humeral circumflex, which enters the head medially at the calcar through intact periosteum (a minor contribution comes from the posterolateral branch of the posterior humeral circumflex and from the rotator-cuff insertions). In a valgus-impacted pattern the head is driven into the metaphysis and the medial periosteal hinge stays intact, so the arcuate entry is preserved and AVN risk is only 15-25 percent; in a displaced 4-part the hinge is disrupted and AVN risk rises to 75-90 percent. This is why fragment number alone does not set head viability — the medial hinge and calcar length do, the central message of the Hertel perfusion study. The axillary nerve in the field. The axillary nerve arises from the posterior cord (C5-C6), exits the quadrangular space with the posterior humeral circumflex artery, and its anterior division runs along the inferior border of subscapularis at the level of the surgical neck, 5-7 cm distal to the acromion, innervating deltoid (motor), teres minor (motor) and the lateral shoulder skin (the regimental badge area). It is the most at-risk nerve, injured in up to 40 percent of fractures at presentation and in 2-3 percent iatrogenically.

Key evidence on biology and technique. Hertel (JSES 2004) showed in an intra-operative perfusion study of 100 intracapsular fractures that the best predictors of head ischaemia were a short calcar (metaphyseal head extension under 8 mm) and a disrupted medial hinge — the basis for plating the valgus-impacted 4-part. Gardner (J Orthop Trauma 2007) demonstrated that adequate medial support, particularly an inferomedial oblique calcar screw, limited head-height loss to 1.2 mm versus 5.8 mm without it — the evidence behind the calcar screw as the most important screw. Boileau (JSES 2002) showed that tuberosity malposition (final malposition in 50 percent) drives poor results after fracture hemiarthroplasty, which is why tuberosity reconstruction and the move to reverse arthroplasty matter. On the operative-versus-non-operative question, the PROFHER trial and the Cochrane review found no superiority of surgery for most displaced surgical-neck fractures, while Olerud (JSES 2011) found no functional benefit but a 30 percent surgical complication rate for displaced 3-part fractures in the elderly.