Shoulder Arthroplasty Revision

You revise a shoulder arthroplasty for one of several distinct failure modes. Establishing which one — and excluding infection before assuming an aseptic cause — determines the entire operation. The four commonest modes, with their typical share of revision work, are:

Two further indications complete the picture. Rotator-cuff failure presents with progressive superior migration (loss of the acromiohumeral interval), subscapularis rupture with internal-rotation loss, or pseudoparalysis, and it mandates conversion to a reverse. Unexplained pain with well-fixed components is the hardest indication and demands careful patient selection — first exclude infection, nerve injury (EMG), cervical spine pathology (MRI neck) and complex regional pain syndrome, because outcomes are poor when no mechanical cause is found. ### Mandatory preoperative workup Every revision gets the same complete workup. Skipping any component is how infections and bone-loss surprises derail the case. Rule out infection (all cases). Check serum ESR and CRP, knowing they are elevated in only about half of infections and are often normal in low-grade disease — a normal result does not exclude infection. Aspirate the joint whenever there is any suspicion, and interpret it against shoulder-specific thresholds:

Assess bone stock with CT (mandatory). A CT of the shoulder with three-dimensional reconstruction (thin cuts under 2 mm) defines the glenoid bone-loss pattern (central, peripheral or combined), classifies the deficiency as mild (less than 10 mm depth), moderate (10 to 15 mm) or severe (greater than 15 mm or medial-wall breach), and evaluates the humeral canal diameter, cortical thickness and any prior cement mantle. Plan component sizes and the grafting strategy from this scan — radiographs alone are inadequate. Assess the soft tissues. An MRI or ultrasound evaluates rotator-cuff integrity (subscapularis, supraspinatus, infraspinatus); a deficient cuff pushes the decision toward a reverse. Confirm deltoid function and axillary-nerve integrity, and assess passive range under anaesthetic for adhesive capsulitis. Obtain the previous operative records. Request all prior notes and implant stickers to document the previous approach, component types and sizes, any complications and any infection history — this drives the extraction plan. Optimise the patient. Optimise bone health (calcium, vitamin D, consider a DEXA scan), mandate smoking cessation, achieve diabetic control (HbA1c less than 7 percent), and check nutritional status (albumin greater than 3.5 g/dL, lymphocyte count greater than 1,500) and cardiac risk in the elderly or comorbid. Consent specifically for the higher revision risks: nerve injury (3 to 8 percent), infection (5 to 15 percent in revision), periprosthetic fracture, instability, stiffness, and re-revision of 10 to 20 percent at five years — with pain relief and range of motion lower than after a primary. Counsel that deltoid or axillary-nerve injury is catastrophic because a reverse cannot function without the deltoid.

The goal is to revise the failed construct through an extensile deltopectoral exposure that protects the deltoid and the axillary nerve, remove the components without sacrificing bone (using a humeral window or extended osteotomy for well-fixed cement), sample for occult infection, reconstruct the bone deficiency, and most often convert to a reverse — which is more forgiving of cuff failure and bone loss. The exposure is laid out in full as the first steps below, and in depth on the deltopectoral approach to the shoulder page.

View larger

Rehabilitation | Phase | Timing | Immobilisation | Rehabilitation focus |

|-------|--------|----------------|----------------------| | 1 | 0–6 weeks | Sling (6–8 weeks if osteotomy or extensive graft) | Passive range only, therapist-assisted; no active motion after a subscapularis repair or osteotomy | | 2 | 6–12 weeks | Sling for heavy tasks only | Active-assisted range once subscapularis healing is confirmed | | 3 | 12 weeks onward | None | Active range and progressive strengthening (deferred to 16–20 weeks after an osteotomy or major graft) | | Protected osteotomy | First 6 weeks | — | No active elbow flexion, to protect a humeral window or extended osteotomy | Recovery is longer than a primary: 6 to 12 months to plateau, with range-of-motion expectations lower. Set realistic patient expectations early. ### Major complications

Epidemiology and global practice. Shoulder arthroplasty volumes are rising worldwide. The National Joint Registry (UK), American Joint Replacement Registry (USA) and Australian Orthopaedic Association National Joint Replacement Registry all report reverse designs as the dominant primary and revision construct, with instability and infection the leading early revision causes. Re-revision runs at 10 to 20 percent at five years (versus 2 to 4 percent for a primary), and outcomes fall with each successive revision — which is why preventing primary failure matters more than perfecting revision technique. Glenoid bone-loss classification (Antuna). The most widely used severity grading guides reconstruction:

Glenoid morphology (modified Walch, Bercik 2016 update). Used in primary glenohumeral osteoarthritis to guide version correction and bone-graft or augment planning, and it informs how much asymmetry a revision construct must accommodate:

The consistent evidence message. A reverse reliably salvages the failed anatomic arthroplasty in the cuff-deficient or bone-deficient shoulder (Patel), though roughly one in six outcomes remain unsatisfactory and revision results are inferior to a primary. Where glenoid bone stock allows, reimplanting a glenoid or baseplate outperforms simple component removal — the basis for liberal bone grafting and augmented or staged reconstruction (Antuna). When a reverse itself fails, instability rather than glenoid loosening dominates, and restoring humeral length and lateralisation is the key technical lever (Boileau). And because Cutibacterium acnes is indolent and mimics aseptic loosening, multiple deep cultures held for at least 14 days are mandatory in every revision, especially in younger men (Singh).