total shoulder arthroplast allPoly Glenoid












Paralysis of deltoid

Torn rotator cuff

Insufficient glenoid bone stock




1.  Functioning / repairable rotator cuff

- maintain stability

- maintain centre of rotation

- early failure if cuff deficient due to rocking horse effect on glenoid component


TSR plus cuff repair

Repaired rotator cuff


Shoulder OAShoulder OA intact cuff on MRI

MRI demonstrating intact cuff


2.  Intact Deltoid


3.  Reasonable glenoid bone stock

- commonly posterior bone loss in OA

- glenoid component must be completely supported by peripheral bone rim or early failure






Shoulder CT Glenoid OvalShoulder Glenoid CT Pear ShapedTriangular vault scapulaGlenoid Inclination

Pear shaped                             Elliptical                                     Vault                                        Superior inclination


Alashkam et al. Clin Anatomy 2021

- approximately 70% glenoids are pear shaped due to presence of glenoid notch anterior margin

- remainder elliptical

- glenoid height varies 30 - 50 mm (mean 35 - 40)

- glenoid width varies 17 - 40 mm (mean 22 - 30 mm)

- inclination superior 4 - 5o




Matsumura et al. JSES 2014

- CT scan 410 normal shoulders

- mean retroversion 1o +/- 3o

- wide range -9o to 13o

- increased glenoid retroversion in men > women, and in dominant arm


Humeral head and scapula


Scapula 30o anteverted from coronal plane


Matsumura et al. JSES 2014

- humeral head retroversion mean 26o +/- 11o

- wide range -2 - 60o

- increased glenoid and humeral (both) retroversion in men > women, and in dominant arm


Glenohumeral OA Classification




Glenohumeral joint OA Grade 2Type 3 Glenohumeral joint OA

Grade 2                                           Grade 3


Grade 1: Osteophytes < 3 mm

Grade 2: Osteophytes 3 - 7 mm with slight narrowing GHJ

Grade 3: Ostephytes > 7 mm with GHJ narrowing and sclerosis


Glenoid morphology, bone stock and version




1.  What is the morphology?

2.  Is there sufficient bone stock for glenoid replacement?

3.  Does version need to be addressed?


Walch classification glenoid morphology


Walch Glenoid Morphology Classfication


Type A - centred humeral head, concentric wear, no humeral head subluxation

A1: minor central erosion

A2: major central erosion with humeral head protrusion


Type B - posterior subluxation of the humeral head, with biconcave glenoid and asymmetric wear

B1: narrowing of the posterior joint space

B2: biconcave glenoid with posterior rim erosion and retroverted glenoid

B3: monoconcave glenoid with > 15° retroversion or >70% posterior humeral head subluxation or both


Type C

C1: dysplastic glenoid with >25° retroversion

C2: biconcave, posterior bone loss, posterior translation of the humeral head


Type D: glenoid anteversion or anterior humeral head subluxation <40°


Type A2 glenoidType B2 glenoid

Type A2                                                   Type B1


Type B2 glenoidScapula CT for total shoulder

Type B2                                                  Type B3


Type B3

Type B3


Accuracy of Walch classification


Shukla et al. JSES 2019

- xray and CT of 100 shoulders with GHJ OA

- intra-observer reliability for xray and CT substantial (0.73)

- inter-observer reliability for xray and CT moderate (0.55, 0.52)


Lowe et al. JSES 2017

- CT v MRI

- largely comparable

- MRI less accurate at distinguishing between type B2 and type C


Progression over time


Walker et al. JBJS Am 2018

- 65 patients with shoulder OA with CT scans 2 years apart

- 8/42 type A progressed

- 17/19 type B1 progressed


Glenoid bone stock


CT TSR Preop Glenoid CystsRheumatoid shoulder CTType A1 glenoid


Glenoid version measurement


1. Xray v CT


Nyffeler et al. JSES 2003

- glenoid version axillary xray v CT

- glenoid retroversion overestimated on xray in 86%

- mean difference in measurements between xray and CT was 6.5o


2. Include scapula


Chalmers et al. J Should Elbow Surg 2017

- 14 B2 glenoids

- glenoid version measurement accurate if > 50% of the scapula width included


3. Measurement techniques


a. Friedman method

- tip of the medial border of the scapula to the midpoint of the glenoid fossa


Scapula CT for total shoulderFriedman method glenoid version


b. Scapula body method

- intersection of the scapula body axis and the glenoid surface


Type B2 glenoidScapula body method glenoid version


Roleau et al. JSES 2019

- 3o difference in measurement of glenoid version between the two

- excellent reliability for both measurement techniques


c. Glenoid vault method

- tip of scapula vault to centre of glenoid


Type B2 glenoidGlenoid vault version measurement


Automated 3D measurement of version


Boileau et al. JBJS Am 2018

- automated software 3D measurement versus 2D Friedman method in 60 shoulders

- mean difference of 2o

- 3D software reliable and accurate


3D CT glenoid 13D CT glenoid 2


Anatomical Total Shoulder Design






Glenoid component design

Glenoid position

Glenoid version




Early highly constrained had high failure rates


Modern design

- unconstrained

- high levels of glenohumeral mismatch and high levels of humeral head translation

- improved ROM and reduces rim stresses on glenoid


Semi Constrained

- cupped glenoid





1. Long stems / short stems / resurfacing

2. Cemented versus uncemented


TSR Uncemented StemTSR Cemented Stem


Stemmed humeral implants


Australian Joint Registry 2021

- 8324 aTSA with stemmed humeral implants

- 10 year revision rate 12%

- 14 year revision rate 15%


Cemented versus uncemented humeral implants


Werthal et al. Bone Joint J 2017

- 4636 shoulders (1167 cemented and 3469 uncemented)

- survival without loosening at 20 years 98% for cemented

- survival without loosening at 20 years 92% uncemented


Short stem


Short stem aTSAShort stem TSA 2


Erickson et al. JSES Int 2020

- systematic review of short stem TSA

- 13 studies with 823 shoulders at 33 month follow up

- 92% uncemented

- 2% humeral loosening, 1% revision for humeral loosening, 3% overall revision rate




TSR No Stem


Willems et al EFORT Open Rev 2021

- systematic review and meta-analysis of 31 articles and 1944 stemless implants

- reduced intra-operative humeral fractures with stemless


Wiater et al. JBJS Am 2021

- RCT of stemless and stemmed implants in 265 shoulders

- no difference in outcomes or complications


Glenoid design


Glenoid options


All poly v metal back

Peg v keel

Flat versus curved back


All poly v metal back


TSR Metal Backed Glenoid


Uncemented metal back advantage

- secured with screws

- can convert to reverse easily if needed


Uncemented metal back disadvantage

- thicker - issues with overstuffing

- issues with metal / poly interface

- higher rate of revision


Page et al. JSES 2018

- Australian Joint Registry

- 11,000 aTSA at 5 years follow up

- revision rate cemented glenoid 3.7% v 17.9% uncemented

- most common revision reasons uncemented were rotator cuff insufficiency (4.4% v 0.4%) and instability (3.8% v 0.8%)

- revision for loosening 1.1% for both groups


Peg v keel cemented all poly


Keeled glenoid polyPegged glenoid poly


Welsher et al. JSES All Access 2019

- meta-analysis of comparative studies

- higher rate of revision with keel v peg

- no difference in functional outcomes


Glenoid positioning


Standard glenoid instrumentationPSI glenoid

Standard                               PSI



- standard instrumentation based upon identifying the glenoid centre point

- navigation

- PSI - patient specific instrumentation


Burns et al. Should Elbow 2019

- comparison standard v navigation v PSI

- both navigation and PSI improved glenoid positioning

- high rate glenoid malposition with standard instrumentation


Type B2 glenoids




1.  Eccentric reaming

2.  Reduce humeral head retroversion

3.  Bone grafting

4.  Augmented glenoids


Eccentric reaming


Limit to 10o

- compromises anterior bone stock

- risk peg penetration through vault and glenoid fracture

- associated with glenoid loosening


Walch et al JSES 2011

- excessive glenoid reaming associated with radiological loosening of glenoid


Glenoid bone graft



- graft resorption / nonunion / migration / loss of fixation


Zhang et al Should Elbow 2021

- systematic review of glenoid bone grafting in aTSR

- 7 articles

- revision rate 5.4% at 6 years


Augmented glenoids


Augmented glenoids


Types glenoid augments

Posterior-wedged augment               Stepped augment                    Full wedged augment



- good version correction

- preserves bone stock


Richetti et al. JSES 2021

- CT evaluation of standard glenoid v posterior stepped augmented glenoid

- posterior stepped glenoid better at correcting version in type B2 and B3

- difficult to restore ML joint line position with both with central glenoid erosion i.e. type A2 or B3


Sheth et al. Should Elbow 2022

- systematic review of augmented glenoid components

- 9 studies and 312 patients

- good version correction

- increased radiolucency seen with 16 wedges and 5 mm steps


Outcomes anatomical TSA




Australian Joint Registry 2021

- 10 year revision rate 12%

- 14 year revision rate 15%

- rotator cuff insufficiency most common indication for revision

- followed by instability / loosening / infection / fracture


Revision reasons aTSARevision aTSA


Revision by age


Australian Joint Registry 2021

- 10 year revision rates

- < 55 14%

- 55 - 64 16.5%

- 65 - 74 11%

- > 74 8%


aTSA v revTSA for OA


Parada et al. J Should Elbow Surg 2021

- 2224 aTSA complication rate 11%, revision rate 5.6%

- aTSA complications: 3% cuff failure, 2.5% aseptic glenoid loosening, 1.3% infection rate

- 4158 revTSA complication rate 9%, revision rate 2.5%

- revTSA complications: 2.5% acromial/scapular fracture, 1.4% instability, 1.2% pain