Addressing Glenoid Deformity in Anatomic Total Shoulder Arthroplasty

Addressing Glenoid Deformity in Anatomic Total Shoulder Arthroplasty

Jared M. Mahylis, MD

Eric T. Ricchetti, MD

Joseph P. Iannotti, MD, PhD


Surgical treatment of the glenoid remains the most challenging portion of anatomic total shoulder arthroplasty (ATSA). A spectrum of glenoid deformity can be encountered, and each morphologic pattern presents its own surgical obstacles. This chapter will highlight the surgical treatment options for glenoid deformity in the setting of ATSA.



Glenohumeral (GH) osteoarthritis (OA) exhibits unique eccentric wear patterns in contrast to OA in other large joints.1,2,3 Walch et al were the first to formally classify these patterns in GH OA. This classification did have notable weaknesses due to poor reproducibility4,5 and has since undergone modifications.6,7 Bercik et al expanded the classification system to include new types of B3 glenoid and D glenoids (FIGURE 14.1). Using three-dimensional computed tomography (3D-CT), they described the B3 glenoid as a progression from a B2 glenoid with continued posterior glenoid wear but loss of biconcavity (monoconcave) with at least 15° of retroversion and/or at least 70% posterior humeral head subluxation relative to the scapula. The C glenoid was further defined as having at least 25° of retroversion not caused by posterior erosion to prevent incorrect classification as a more severe B glenoid.6 Subsequently, Iannotti et al further defined the B3 glenoid and added a new subtype, the C2 glenoid. Utilizing 3D-CT with the previously validated 3D glenoid vault model, the B3 glenoid definition was expanded to having minimal or no paleoglenoid, high retroversion due to posterior wear, and more medialization than B2 glenoids (FIGURE 14.2A). The C2 glenoid was introduced as a biconcave variant similar to a B2 glenoid, but with underlying glenoid dysplasia and thus a greater premorbid glenoid retroversion. The new C2 glenoid also exhibited greater posterior humeral head subluxation compared with C1 glenoids resulting from the “acquired” biconcavity (FIGURE 14.2B).7

Imaging Assessment of the Glenoid

Assessment of GH pathology is important to surgically addressing deformity. However, radiographs have shown poor reproducibility and frequently overestimate glenoid retroversion.8 Studies have shown that x-rays demonstrate fair to good assessment of glenoid morphology and Walch classification when compared to CT.9,10 However, intra- and interreader reliability remains only fair to moderate when comparing standard radiographs to more advanced imaging such as CT and magnetic resonance imaging (MRI).10,11

CT remains the standard for assessment of glenoid pathology.12,13,14 Two-dimensional CT has shown fair reproducibility in assessing retroversion and classifying morphology,2,4,5 but with development of 3D-CT reconstruction and surgical planning software, multiple studies have shown superiority of 3D-CT in assessing glenoid morphology.13,15,16 MRI studies assessing glenoid architecture remain mixed.17,18 Raymond et al showed superiority of MRI to axillary radiographs in assessment of retroversion.17 However, Lowe et al found MRI had poor accuracy in assessing the advanced glenoid bone loss and retroversion of B2 and C glenoids.18 Overall, 3D-CT provides a more detailed evaluation of the glenoid in OA.12

Humeral Head Relationship to Glenoid and Scapula in OA

Neer first noted posterior humeral head subluxation with asymmetric GH OA,19 which Walch et al later detailed.2 Walch described a humeral head subluxation index, relative to the center of the glenoid, between 45% and 55% to be a centered humeral head, while 0% was considered an anterior dislocation and 100% a posterior dislocation.2 This study, however, did not compare subluxation to normal controls.

Recently, Sabesan et al utilized 3D-CT comparing 60 patients with advanced OA to 15 non-OA controls to define the baseline relationship between the center of the humeral head to the glenoid fossa plane and to the plane of the scapula (FIGURE 14.3).20 Humeral-scapular alignment (HSA) averaged -8.43 ± 5.58 mm or -17.1% ± 11.3% relative to humeral head diameter in OA patients, compared to -0.998 ± 1.85 mm or -2.27% ± 4.18% in
controls. A near-perfect linear relationship between glenoid retroversion and HSA (P < 0.001) was seen, as well as a significant relationship between glenoid bone loss and HSA (P < 0.001). Humeral-glenoid alignment (HGA) averaged -1.16 ± 3.81 mm or -2.48% ± 7.3% relative to humeral head diameter in OA patients, compared to 0.48 ± 1.33 mm or 0.87% ± 2.6% in controls. The relationship between glenoid retroversion and HGA was not significant, but the relationship between glenoid bone loss and HGA was (P < 0.001). The authors concluded that the two methods of defining humeral head subluxation are different and independent of one another. While HSA is highly correlated with glenoid retroversion, HGA is more variable and related to factors other than bony glenoid anatomy that may include soft-tissue contracture or humeral head deformity.20

Chan et al. showed similar findings when assessing the B3 glenoid. They found a mean posterior humeral head subluxation of 80% ± 8% for HSA, while HGA showed a mean posterior humeral head subluxation of 54% ± 6%. The authors suggested a centering of the humeral head with respect to the neoglenoid articulation. They found a significant correlation between glenoid version and degree of posterior humeral head subluxation related to HSA, with every 1° increase in glenoid retroversion resulting in a 1% increase in posterior humeral head subluxation (P < 0.001), but no significant correlation with HGA and glenoid retroversion (P = 0.409).21

Gender Specifics in OA

Epidemiologic studies show high prevalence of OA,22,23,24 with men and women having different rates of OA,25,26 which does include the shoulder. Nakagawa et al found significantly greater proportion of primary GH OA in female patients compared to males with shoulder disease (P = 0.0178).26 The differences in prevalence and onset between men and women also appear to be different with women having greater prevalence and later onset.26,27 Schoenfeldt et al found women to be affected in 54% of cases of primary GH OA, with an older age at time of presentation for treatment compared to men (72.9 ± 11.2 vs 66.1 ± 12.4; P = 0.0005).27 Lastly, multiple studies have shown asymmetric glenoid wear to be far more prevalent in males, with four to five times greater incidence of B2 and B3 glenoid morphology in males.28,29,30

Natural History and Progression of Disease

The pathogenesis and pathophysiology of GH OA remains poorly understood.1 Historically, primary GH OA was thought to begin on the glenoid.1,2 Yet, few studies have assessed pathologic progression of disease looking at bony and soft-tissue changes in primary OA.31 Walker et al recently performed a retrospective CT assessment of 65 shoulders with GH OA analyzing the amount and location of glenoid bone loss on at least two CT scans performed 24 months apart. Rotator cuff fatty infiltration was calculated and assignment of modified Walch classification was performed on all scans. Of the 65 shoulders assessed, 42 were A-type glenoids and 23 B-type glenoids. Successive CT scans showed 8 of 42 A1 glenoids had evidence of pathologic progression compared to 17 of 19 B1 glenoids (15 to B2 and 2 to B3) (P < 0.001) (FIGURE 14.4A-C). The odds of progressive joint-line medialization were 8.1 times higher for B-type glenoids. B-type glenoids showed a higher association with percentage of infraspinatus muscle fatty infiltration compared with A-type glenoids both initially (14% vs 7%; P < 0.001) and at final CT follow-up (16% vs 10%; P = 0.003). The authors concluded progressive asymmetric bone loss seldomly develops in A1 glenoids, whereas initial posterior humeral head translation (B1 glenoids) is associated with subsequent development and progression of posterior glenoid bone loss.28

Donohue et al further examined the relationship of glenoid morphology and rotator cuff fatty infiltration in 175 patients who underwent TSA for GH OA. Using 3D-CT analysis, glenoid pathologic version, and joint line, modified Walch classification and Goutallier classification were determined. High-grade posterior rotator cuff fatty infiltration (combined infraspinatus and teres minor) was seen in 16% B2 and 55% of B3 glenoids compared to 8% and 12% for A1 and A2 glenoids, respectively (P < 0.001). Higher fatty infiltration of the infraspinatus, teres minor, and combined posterior rotator cuff muscles was associated with increasing glenoid retroversion (P < 0.05), while higher fatty infiltration of all four rotator cuff muscles and combined posterior rotator cuff muscles was associated with increasing joint-line medialization (P < 0.05).29


Correction of Anatomy: Eccentric Reaming and Use of a Standard Glenoid Component

Eccentric or asymmetric reaming (AR) with use of a standard glenoid component currently remains the most common method of addressing glenoid asymmetry (eg, B2 glenoid) in ATSA (FIGURE 14.5A). Glenoid asymmetry such as biconcavity is corrected by reaming the paleoglenoid. Using this technique, surgeons are able to treat approximately 5 to 8 mm of posterior bone loss while also being able to correct version between 10° and 15°.32 This does lead to joint-line medialization (FIGURE 14.5B), which requires greater implant thickness to make up for this bone loss.33 Good clinical outcomes have been reported with this method,34,35,36 although they are limited by study design and duration of follow-up.

The technique does have limitations. Correcting retroversion greater than 15° to 20° results in excessive
joint-line medialization that is associated with glenoid implant perforation,33,37 loss of cortical bone support, and narrowed glenoid dimension that may impact implant stability.38,39 Sabesan et al suggested correcting glenoid retroversion to 6° rather than 0° in order to minimize joint-line medialization and best restore normal anatomy.40

Correction of Anatomy: Augmented Glenoid Component

Current Available Implants

Augmented glenoid components allow for better bone preservation, restoration of the joint line, and better soft-tissue tensioning compared to asymmetric glenoid reaming. These implants also eliminate the need for bone grafting. An early metal-backed wedged design (Smith & Nephew; London, UK) was plagued by similar failure rates as other metal-backed implants. Cil et al reported a 31% 10-year revision-free survival rate of this implant.41 Three posteriorly augmented glenoid components are currently available in the United States, a stepped design (StepTech; Depuy Synthes, Warsaw, Indiana; FIGURE 14.6A), a full-wedge design (Equinoxe Posterior Augment; Exactech, Gainesville, Florida; FIGURE 14.6B), and a posterior wedge design (Aequalis Perform+, Wright Medical Group N.V., Memphis, Tennessee; FIGURE 14.6C).42

Surgical Technique Differences Between Implants

All currently available implants attempt to limit the bone loss associated with AR while correcting or maintaining the native joint line. Each implant manufacturer’s surgical technique should be utilized for detailed steps on correct sizing and insertion of the implant. The following description highlights differences between implants and should not be used in lieu of the manufacturers recommended technique. The shape of the implant augment may be important in determining its best use between a biconcave (B2) glenoid and a glenoid with more symmetric posterior glenoid bone loss (B3). All three manufacturers utilize preoperative planning 3D-CT software to assist with determining the optimal implant position at the time of surgery (TrueMatch, Depuy Synthes, Warsaw, Indiana; GPS, Exactech, Gainesville, Florida; and BluePrint, Wright Medical Group N.V., Memphis, Tennessee).42,43,44

Only gold members can continue reading. Log In or Register to continue

Jun 23, 2022 | Posted by in ORTHOPEDIC | Comments Off on Addressing Glenoid Deformity in Anatomic Total Shoulder Arthroplasty
Premium Wordpress Themes by UFO Themes