Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Literature Review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Search in posts
Search in pages
Filter by Categories
Case Report
Clinical research study
Current Issue
Literature Review
Original Article
Research Article
Review Article
Short Report
Surgical techniques
View/Download PDF

Translate this page into:

76 (); 230-238
doi:
10.1016/j.jor.2026.04.003

Outcomes of anatomic total shoulder arthroplasty in B2 versus B3 glenoid morphology: A systematic review

, , , , , , , ,
a Virginia Commonwealth University School of Medicine, PO BOX 980146, Richmond, VA, 23298-0146, USA
b Department of Orthopaedic Surgery, Virginia Commonwealth University Medical Center, 1200 E Broad St., 9th Floor, P.O. Box 980153, Richmond, VA, 23298, USA

⁎Corresponding author: Zachary Whong. whongz@vcu.edu

Received: , Accepted: ,
Professor P K Surendran Memorial Education Foundation
Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Abstract

Glenohumeral osteoarthritis (GHOA) with Walch B2 or B3 glenoid morphology poses challenges during anatomic total shoulder arthroplasty (aTSA) due to posterior wear and retroversion. While B2 glenoids are well studied, outcomes for B3 glenoids remain unclear. This systematic review compared aTSA outcomes between B2 and B3 glenoids.

Following PRISMA guidelines, PubMed, Embase, and Cochrane CENTRAL were searched through March 2025 for studies comparing aTSA outcomes in B2 versus B3 glenoids. Extracted data included demographics, patient-reported outcome measures (PROMs), range of motion (ROM), and radiographic findings. Risk of bias was assessed using JBI Critical Appraisal Tools.

Six studies (443 patients; B2 = 290, B3 = 153) were included. Both groups showed significant postoperative improvement in PROMs and pain. ROM improved in both, though gains in flexion and abduction were nonsignificant for B3 glenoids. Internal rotation improved more in B3 glenoids with stepped augmented components. Radiographically, both showed correction of version and inclination, but B3 glenoids remained more medialized. Radiolucency was more frequent in B2 (31%) than B3 (22%) glenoids.

aTSA is associated with meaningful functional and radiographic improvement in both B2 and B3 glenoids. While improvements are observed across deformity types, direct comparisons between B2 and B3 outcomes are limited, and persistent joint line medialization in B3 glenoids highlights the ongoing challenges of managing more severe deformity.

Keywords

Glenohumeral osteoarthritis
Anatomic total shoulder arthroplasty
Walch classification
B2 glenoid
B3 glenoid
Posterior glenoid wear
Functional outcomes
Range of motion
1

1 Introduction

Glenohumeral osteoarthritis (GHOA) affects nearly 20% of adults over the age of 65 and comprises up to 17% of all shoulder complaints.1,2 Anatomic total shoulder arthroplasty (aTSA) is the gold standard for GHOA in patients with an intact rotator cuff. Over the past decade, the number of aTSAs performed annually has increased nearly 1.5-fold, with an estimated 122% increase by the year 2040.3,4 While aTSA provides excellent pain relief and functional results for most patients, those with glenoid bone loss, deformity, or rotator cuff dysfunction are more susceptible to post-procedural complications.5–7.

To better prognosticate risk of adverse outcomes based on glenoid morphology, Walch et al. developed a classification system describing different glenoid subtypes.8 In this system, type B glenoids (B1 and B2) are characterized by retroversion, biconcavity, and posterior humeral head subluxation.9 Bercik et al. later added a B3 subtype, defined as uniconcave with ≥15° of retroversion and/or 70% posterior humeral head subluxation.10 More recently, Ianotti et al. added granularity to the Walch subtypes by using 3D CT to better delineate glenoid morphologic features. They determined that B3 glenoids had significantly greater joint line medialization and a more centered humeral head than B2 glenoids; and furthermore, that B2 glenoids may progress to B3 glenoids with continued posterior erosion.11

The posterior erosion and humeral head subluxation inherent to B2 and B3 glenoids pose unique challenges during aTSA.12 While implications of the B2 subtype are well described, the B3 subtype has received comparatively less attention. Many studies group these two subtypes together and few compare them. This represents a critical oversight as differences in surgical outcomes between B2 and B3 subtypes may influence future implant design and surgical decision-making. This systematic review aims to compare the outcomes of patients with B2 and B3 glenoids undergoing aTSA for treatment of GHOA.

2

2 Methods

2.1

2.1 Literature search

A literature search was conducted through March 24, 2025 using the Embase (Ovid), Pubmed/MEDLINE, and Cochrane CENTRAL databases. The following search terms were used to narrow down articles relative to the area of interest: (((“Arthroplasty, Replacement" [Mesh] AND “Shoulder Joint" [Mesh]) OR “Arthroplasty, Replacement, Shoulder" [Mesh] OR “Shoulder Replacement" [Title/abstract:∼2] OR “Shoulder Replacements" [Title/abstract:∼2] OR “Shoulder Arthroplasty" [Title/abstract:∼2] OR “Shoulder Arthroplasties" [Title/abstract:∼2]) AND (“Glenoid Cavity" [Mesh] OR glenoid)) AND (wear OR Walch [tiab] OR B2 OR B3). The specific search strategies for each database are summarized in Appendix Tables A.1, A.2, and A.3. After relevant literature was queried, studies were exported to Covidence for screening and data extraction. The systematic review followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting standards.

2.2

2.2 Literature screening

Studies exported to Covidence underwent two rounds of screening by two independent reviewers. Inclusion criteria were studies involving patients with B2 or B3 glenoid morphology who underwent aTSA and reported post-operative outcomes stratified by Walch classification. Systematic reviews, case reports, technique papers, editorials, or conference abstracts were excluded. Additional exclusion criteria included studies that involved cadavers, animals, biomechanics, were not published in English, or were unavailable as a full text. The first round of screening involved a review of titles and abstracts, followed by full-text review of the same eligibility criteria. Screening was conducted by Z.W. and A.K.C. with conflicts resolved by S.W.

2.3

2.3 Risk of bias

Methodological strength of included research articles underwent risk of bias review according to JBI's Critical Appraisal Tool. Risk of bias review was conducted by Z.W. and A.K.C. with conflicts resolved by S.W. All six studies were deemed eligible for inclusion.

2.4

2.4 Data extraction

Studies that passed both rounds of screening and qualified for inclusion were advanced for data extraction. Key variables recorded included demographic information, Walch classification, implant details, follow-up duration, patient-reported outcome measures (PROMs), range of motion, radiographic findings, and pain scores. Patients were only included if follow-up was available. Data extraction was conducted independently by Z.W. and A.K.C. followed by a second round of verification to ensure accuracy of information. Due to the heterogeneity of reported patient outcomes, a meta-analysis was not performed.

3

3 Results

3.1

3.1 Database screening

The initial search yielded a total of 1124 articles. After removing duplicate articles, 697 studies underwent title and abstract screening, 40 of which were advanced to full-text screening. Thirty-four articles were subsequently excluded due to the following reasons: abstract-only publication (n = 1), unavailable full-text (n = 4), incorrect outcomes (n = 8), incorrect intervention (n = 3), or incorrect study design (n = 18). Final data extraction was performed on 6 articles (Fig. 1).

Flow chart of article screening based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Guidelines.
Fig. 1 Flow chart of article screening based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Guidelines.
3.2

3.2 Patient demographics

A total of 443 patients underwent aTSA for the treatment of glenohumeral osteoarthritis. 65% (n = 290) presented with B2 glenoid wear and 35% (n = 153) presented with B3 glenoid wear (Table 1). Sex distribution was reported for 383 patients; among these, 61% (n = 233) were male and 39% (n = 150) were female. The weighted average age of the entire patient population was 67.6 years. Multiple studies reported sex distribution and average age only for the overall cohort, which included other glenoid subtypes, and therefore could not be incorporated into the subtype-specific sex analysis.

Table 1 Study characteristics.
Study Study Design N (male/female) Mean Age (SD; Range) B2 (N) B3 (N) Implant Type Follow-up Duration Outcomes Reported Notes
Magosch 2021 Retrospective cohort study 48 (28/20) (Full cohort) ∗ 67.3 (range 46-77 years) 14 6 Universal Glenoid Component™ 49 (range 24-77 months) Active flexion, external rotation, relative constant Murley score, absolute constant Murley score, constant Murley score pain/ADL/ROM/strength ∗Sex not reported by glenoid type.
Polisetty 2023 Retrospective cohort study 101 (54/47) (Full cohort) ∗ 71 ± 6.3 years 70 31 Standard cemented all-polyethylene glenoid component 47 (range 24-122 months) Active flexion, internal rotation, external rotation, VAS pain score, SANE score, presence/absence of radiolucency, achievement of MCID/SCB Comparison of rTSA and TSA.∗Sex not reported by glenoid type.
Ianotti 2021 Prospective cohort study B2: 29 (20/9)B3: 20 (17/3) B2: 61.8 ± 5.8B3: 65.9 ± 6.5 29 20 Global Anchor Peg Glenoid; Depuy Synthes 2.3 ± 0.4 years Penn shoulder score, glenoid version inclination, joint line relative to vault, HSA-AP, HGA-AP Global STEPTECH Anchor Peg Glenoid, PREMIERON X-Linked Polyethylene; DePuy Synthes is a posteriorly stepped augmented glenoid component
Gutman 2023 Prospective cohort study B2: 41 (30/11)B3: 9 (8/1) B2: 66.5 ± 7.66B3: 72.3 ± 8.25 41 9 StepTech APG; DePuy Orthopaedics, Warsaw, IN, USA 42 months (range 24-106 months) Charlson Comorbidity Index, forward elevation, external rotation, internal rotation, VAS pain score, SANE score
Ricchetti 2021 Prospective cohort study B2: 53 (SG n = 22, AG n = 31∗)B3: 35 (SG n = 12, AG n = 23∗) 63 ± 8 years (range 43-90 years) 53 35 Global Anchor Peg Glenoid; DePuy Synthes, Raynham, MA, USAGlobal STEPTECH Anchor Peg Glenoid; DePuy Synthes posteriorly augmented stepped polyethylene anchor peg glenoid component Within 3 months of surgery Glenoid version, inclination, ML joint line relative to pathologic and premorbid anatomy SG – standard glenoidAG – posteriorly stepped augmented glenoid
Matsen 2020 Retrospective cohort study B2: 83 (45/38)B3: 52 (31/21) B2: 68 ± 9 (range 42-86 years)B3: 72 ± 8 (range 55-87 years) 83 52 Standard, non-augmented, all polyethylene glenoid component (GLOBAL Anchor Peg; DePuy Synthes)Standard length, impaction-grafted humeral stem (GLOBAL ADVANTAGE; DePuy Synthes) 2.1 ± 0.2 (range 2-3 years) Simple Shoulder Test (SST), percentage of humeral head decentering on glenoid face, bone ingrowth into the central peg Studies outcomes of treatment of type B2 and B3 glenoids using a standard glenoid component inserted with conservative glenoid reaming without normalizing glenoid version.
3.3

3.3 Functional outcomes

One study reported significant improvements in Simple Shoulder Test (SST) scores in both glenoid subtypes (Table 2). In the B2 group, SST improved from 3.0 ± 2.1 to 9.8 ± 2.1 (p < 0.001), while the B3 group saw improvement to 9.8 ± 2.1 from 3.0 ± 2.4 (p < 0.001).13 Another study evaluated Penn Shoulder Scores and found B2 glenoids had a greater increase in mean score compared to B3 glenoids (Table 2).14 The mean increase in the B2 group was 66.8 (preop: 30.6, postop: 97.4) at minimum 2-year follow-up, compared to a 59.6 point increase in the B3 group (preop: 37.1, postop: 96.7) over the same period.14 However, B2 and B3 glenoids were not directly compared, precluding detection of any statistically significant differences.14 Across studies, post-operative ASES scores were significantly improved in both B2 and B3 glenoids, with no significant differences between groups (Table 2).12 Polisetty et al. reported similar post-op ASES scores (B2: 83.6 ± 21.5, B3: 84.3 ± 21.5) and found that achievement of Minimally Clinically Important Difference (B2: 96%, n = 67; B3: 94%, n = 29) and Substantial Clinical Benefit (B2: 81%, n = 57; B3: 81%, n = 25) were high in both groups.12 Absolute and Relative Constant Murley Scores also improved significantly for both B2 and B3 glenoids (Table 2). While B2 glenoids achieved higher post-operative values, this study likewise lacked statistical B2 to B3 comparison.15 Magosch et al. reported improved Absolute and Relative Constant Murley scores in both B2 (Absolute: 41.2 ± 13.7 to 77.9 ± 12.8, P = 0.001) (Relative: 53.8 ± 18.9 to 104.5 ± 23.4, P = 0.001) and B3 glenoids (Absolute: 31.8 ± 16.2 to 69.3 ± 17.1, P = 0.028) (Relative: 40.7 ± 18.4 to 87.5 ± 22.4, P = 0.042) following aTSA with a convertible glenoid component.15 Multiple studies reported significant improvements in SANE scores for both B2 and B3 glenoids, with post-operative means in the mid 80s and low 90s and no major differences between subtype (Table 2).12,16,17

Table 2 Functional outcome scores.
Study Simple Shoulder Test (SST) or Penn Shoulder Score ASESAchievement of MCID/SCB Absolute Constant Murley Score Relative Constant Murley Score SANE
Matsen 2020 SST: N/A N/A N/A
B2 preop: 3.0 ± 2.1 (0-8) B2 preop: 40 ± 19
B2 postop: 9.8 ± 2.1 (1-12) B2 postop: 85 ± 15
P < 0.001 P < 0.001
B3 preop: 3.0 ± 2.4 (0-10) B3 preop: 38 ± 23
B3 postop: 9.8 ± 2.1 (5-12) B3 postop: 86 ± 13
P < 0.001 P < 0.001
Magosch 2021 N/A N/A B2 preop: 41.2 ± 13.7 B2 preop: 53.8 ± 18.9 N/A
B2 postop: 77.9 ± 12.8 B2 postop: 104.5 ± 23.4
P = 0.001 P = 0.001
B3 preop: 31.8 ± 16.2 B3 preop: 40.7 ± 18.4
B3 postop: 69.3 ± 17.1 B3 postop: 87.5 ± 22.4
P = 0.028 P = 0.042
Ianotti 2021 Penn Shoulder Score N/A N/A N/A N/A
A1 preop: 33.1 (range 24.8-42.0)
B2 preop: 30.6 (range 24.0-53.0)
B3 preop: 37.1 (range 19.0-47.0)
A1 minimum 2-year follow-up: 95.0 (range 88.0-99.0)
B2 minimum 2-year follow-up: 97.4 (range 90.0-100)
B3 minimum 2-year follow-up: 96.7 (range 90.0-100)
Gutman 2023 N/A N/A N/A S B2 postop: 95.5 ± 5.17 (range 93.7-97.3)
B3 postop: 90.6 ± 5.83 (range 86.1-95.0)
P = 0.04
Polisetty 2023 N/A B2 postop: 83.6 ± 21.5 N/A N/A B2 postop: 87.7 ± 17.8
B3 postop: 84.3 ± 21.5 B3 postop: 88.8 ± 19.1
B2 MCID: 67 (96%)
B3 MCID: 29 (94%)
B2 SCB: 57 (81%)
B3 SCB: 25 (81%)
3.4

3.4 Range of motion

Both B2 and B3 glenoids demonstrated significant improvement in range of motion (ROM) as measured by Constant Murley Score ROM following aTSA (B2: 18.9 ± 5.6 preop to 32.5 ± 5.7 postop, P = 0.002; B3: 15.3 ± 6.3 preop to 29.7 ± 7.7 postop, P = 0.043) (Table 3).15 While flexion and abduction improved significantly in B2 glenoids, B3 glenoids showed improvement that failed to reach significance (Table 3).12,15 Magosch et al. reported significant improvement in flexion for B2 glenoids (119.2° ± 19.8 preop to 157.7° ± 19.2 postop, P = 0.004) compared with a nonsignificant change in B3 glenoids (100.0° ± 24.5 to 145.0° ± 28.1, P = 0.072).15 Similarly, abduction improved significantly in B2 glenoids (104.6° ± 28.8 to 149.6° ± 21.5, P = 0.007) but not in B3 glenoids (75.0° ± 31.5 to 136.7° ± 32.7, P = 0.08).15

Table 3 Range of motion.
Study Flexion Abduction External Rotation Internal Rotation Forward Elevation Constant Murley Score ROM (points)
Magosch 2021 B2 preop: 119.2 ± 19.8 B2 preop: 104.6 ± 28.8 B2 preop: 18.0 ± 18.0 N/A N/A B2 preop: 18.9 ± 5.6
B2 postop 157.7 ± 19.2 B2 postop 149.6 ± 21.5 B2 postop 46.2 ± 19.4 B2 postop: 32.5 ± 5.7
P = 0.004 P = 0.007 P = 0.003 P = 0.002
B3 preop: 100.0 ± 24.5 B3 preop: 75.0 ± 31.5 B3 preop: 20.0 ± 27.7 B3 preop: 15.3 ± 6.3
B3 postop 145.0 ± 28.1 B3 postop 136.7 ± 32.7 B3 postop 52.5 ± 14.1 B3 postop: 29.7 ± 7.7
P = 0.072 P = 0.08 P = 0.042 P = 0.043
Polisetty 2023 B2 postop: 143 ± 18 N/A B2 postop: 54 ± 19 B2 postop: 6.6 ± 2.1 N/A N/A
B3 postop: 140 ± 17 B3 postop: 55 ± 14 B3 postop: 6.2 ± 2.1
Gutman 2023 N/A N/A B2 preop: 21.9 ± 13.0 (range 17.7-26.0) B2 preop: S1 (L5-S1) B2 preop: 112 ± 27.7 (range 104-121) N/A
B3 preop: 18.3 ± 13.7 (range 7.81-28.9) B3 preop: S1 (L5-S2) B3 preop: 100 ± 24.5 (range 81.2-119)
P = 0.494 P = 0.439 P = 0.2
B2 postop: 41.9 ± 6.57 (range 39.8-44.0) B2 postop: L1 (T12-L2) B2 postop: 156 ± 11.6 (range 152-160)
B3 postop: 40.3 ± 5.34 (range 36.2-44.4) B3 postop: T10 (T8-L1) B3 postop: 150 ± 9.01 (range 143-157)
P = 0.467 P = 0.024 P = 0.103

One study reported improvement in external rotation for B2 glenoids from 18.0° ± 18.0 preop to 46.2° ± 19.4 postop (P = 0.003) and B3 glenoids from 20.0° ± 27.7 to 52.5° ± 14.1 (P = 0.042).15 Another study reported similar postoperative external rotation between groups (B2: 41.9° ± 6.6 [range 39.8–44.0]; B3: 40.3° ± 5.3 [range 36.2–44.4]; P = 0.467).16 Internal rotation improved in both groups, with one study reporting greater improvement in B3 glenoids (Table 3).16 In a study by Gutman et al. mean preoperative internal rotation was S1 (range L5–S1) for B2 glenoids and S1 (range L5–S2) for B3 glenoids, with no significant difference between groups (P = 0.439).16 Postoperatively, internal rotation improved to L1 (range T12–L2) in B2 glenoids and T10 (range T8–L1) in B3 glenoids, with a significant difference favoring B3 glenoids (P = 0.024).16

3.5

3.5 Radiographic outcomes

Preoperatively, B2 glenoids were more posteriorly subluxated than B3 glenoids (B2 Humeral-Scapular Alignment [HSA-AP]: −26.6° ± 10.4; B3 HSA-AP: −19.9° ± 5.6; p < 0.001, measured across A1, B2, and B3).14 Immediately postoperatively, HSA-AP improved in both groups (B2: −17.0° ± 9.5; B3: −11.0° ± 6.6; p < 0.001, across A1, B2, B3) and remained improved at ≥2-year follow-up (B2: −14.6° [–21.4 to −9.4]; B3: −10.3° [–14.0 to −6.9]; p < 0.001, across A1, B2, B3).14 For Humeral-Glenoid Alignment (HGA-AP), B2 glenoids were also more malaligned preoperatively (B2: −8.7° [–12.5 to −6.2]; B3: −1.0° [–3.3 to 0.8]; p < 0.001, across A1, B2, and B3), with immediate postoperative and ≥2-year follow-up values showing no significant differences between subtypes (immediate postop: B2: −0.8° [–1.7 to −0.2]; B3: 0.2° [–0.8 to 1.3]; p = 0.19, across A1, B2, B3; ≥2-year: B2: −0.1° [–1.8 to 2.8]; B3: −0.2° [–2.7 to 1.2]; p = 0.66, across A1, B2, B3) (Table 4).14

Table 4 Radiographic outcomes.
Study Glenoid Version Inclination Joint Line Decentering (%)And Radiolucency Humeral Scapular Alignment (HSA-AP) Humeral Glenoid Alignment (HGA-AP) Presence of Central Peg Osteolysis (CPO)
Matsen 2020 N/A N/A N/A Decentering (%) N/A N/A N/A
B2 preop: −14 ± 7 (range -42-0)
B2 postop: −1 ± 2 (range -7-0)
P < 0.001
B3 preop: −4 ± 6 (range -17-8)
B3 postop: −1 ± 3 (range -9-0)
P < 0.0269
Ianotti 2021 A1 SG preop: −6.1 (range −9.1, −3.2) A1 SG preop: 4.9 ± 4.4 Joint Line Relative to Vault N/A A1 SG preop HSA-AP: −6.5 ± 7.3 A1 SG preop HGA-AP: −1.4 (range −3.9, 1.9) A1 SG: 2 (5%)
B2 AG preop: −20.0 (range −24.5, −15.6) B2 AG preop: 6.2 ± 5.4 A1 SG preop: −1.6 (range −2.2, −1.2) B2 AG preop HSA-AP: −26.6 ± 10.4 B2 AG preop HGA-AP: −8.7 (range −12.5, −6.2) B2 AG: 3 (10%)
B3 AG preop: −22.1 (range −25.4, −17.0) B3 AG preop: 4.4 ± 5.6 B2 AG preop: −2.2 (range −3.3, −0.84) B3 AG preop HSA-AP: −19.9 ± 5.6 B3 AG preop HGA-AP: −1.0 (−3.3, 0.80) B3 AG: 6 (29%)
P < 0.001 P = 0.39 B3 AG preop: −5.2 (range −6.1, −4.3) P < 0.001 P < 0.001 P = 0.028
A1 SG immediate postop: −6.5/- 5.3 A1 SG immediate postop: 6.1 ± 4.1 P < 0.001 A1 SG immediate postop HSA-AP: −6.2 ± 7.2 A1 SG immediate postop HGA-AP: −0.25 (range −2.3, 2.0)
B2 AG immediate postop: −10.3 ± 6.5 B2 AG immediate postop: 6.2 ± 5.3 A1 SG within 3 months postop: 1.00 (range 0.50, 1.6) B2 AG immediate postop HSA-AP: −17.0 ± 9.5 B2 AG immediate postop HGA-AP: −0.80 (range −1.7, −0.20)
B3 AG immediate postop: −8.1 ± 4.7 B3 AG immediate postop: 5.6 ± 4.2 B2 AG within 3 months postop: 0.40 (range −0.30, 1.5) B3 AG immediate postop HSA-AP: −11.0 ± 6.6 B3 AG immediate postop HGA-AP: 0.20 (range −0.80, 1.3)
P = 0.020 P = 0.88 B3 AG within 3 months postop: −1.7 (range −3.1, −0.60) P < 0.001 P = 0.19
A1 SG minimum 2-year follow-up: −6.9 ± 5.4 A1 SG minimum 2-year follow-up: 8.5 ± 4.9 A1 SG minimum 2-year follow-up HSA-AP: −6.7 (range −11.2, 0.80) A1 SG minimum 2-year follow-up HGA-AP: −0.10 (range −2.8, 4.1)
B2 AG minimum 2-year follow-up: B2 AG minimum 2-year follow-up: B2 AG minimum 2-year postop HSA-AP: −14.6 (range −21.4, −9.4) B2 AG minimum 2-year postop HGA-AP: −0.10 (range −1.8, 2.8)
11.1 ± 6.6 8.4 ± 6.6 B3 AG minimum 2-year postop HSA-AP: −10.3 (range −14.0, −6.9) B3 AG minimum 2-year postop HGA-AP: −0.20 (range −2.7, 1.2)
B3 AG minimum 2-year follow-up: B3 AG minimum 2-year follow-up: P < 0.001 P = 0.66
−8.9 ± 5.2 9.6 ± 6.1
P = 0.012 P = 0.72
Ricchetti 2021 B2 AG CT2-CT1: 9.5 ± 6.4 B2 AG CT2-CT1: 0.6 ± 4.6 Glenoid ML Joint Line Position (mm) N/A B2 AG: −17.2 ± 9.3 B2 AG: −1.4 ± 3.8 N/A
B2 AG CT2-Vault: −1.7 ± 6.6 B2 AG CT2-Vault: −0.6 ± 4.7 B2 AG CT2-CT1: 2.5 ± 1.2 B3 AG: −12.3 ± 7.6 B3 AG: −0.50 ± 3.0
B3 AG CT2-CT1: 13.2 ± 7.1 B3 AG CT2-CT1: 1.3 ± 5.1 B2 AG CT2-Vault: 0.3 ± 1.6 P = 0.042 P = 0.33
B3 AG CT2-Vault: −0.6 ± 5.1 B3 AG CT2-Vault: −0.3 ± 4.2 B3 AG CT2-CT1: 3.2 ± 1.4
CT2-CT1 p-values CT2-CT1 p-values B3 AG CT2-Vault: −2.2 ± 2.1
B3 AG vs. B2 AG B3 AG vs. B2 AG CT2-CT1 p-values
P = 0.046 P = 0.60 B3 AG vs. B2 AG
CT2-Vault p-values CT2-Vault p-values P = 0.042
B3 AG vs. B2 AG B3 AG vs. B2 AG CT2-Vault p-values
P = 0.52 P = 0.79 B3 AG vs. B2 AG
P < 0.001
Polisetty 2023 N/A N/A N/A Radiolucency N/A N/A N/A
B2: 22 (31.4%)
B3: 7 (22.6%)

Glenoid version and inclination improved in both groups following aTSA, though B3 glenoids remained more medialized at follow-up.14,17 Mean change in version was 0.6° ± 4.6 for B2 glenoids and 1.3° ± 5.1 for B3 glenoids (p = 0.60). CT-Vault analysis showed similar findings (B2: −0.6° ± 4.7; B3: −0.3° ± 4.2; p = 0.79). Preoperatively, inclination was 6.2° ± 5.4 for B2 and 4.4° ± 5.6 for B3 glenoids. Immediate postoperative values were 6.2° ± 5.3 for B2 and 5.6° ± 4.2 for B3, and at ≥2-year follow-up −8.4° ± 6.6 (B2), and 9.6° ± 6.1 (B3). Differences across glenoid types (including A1 glenoids) were not statistically significant (preoperative p = 0.39, postoperative p = 0.88, ≥2-year follow-up p = 0.72) (Table 4).14,17 Preoperatively, joint lines were more medialized in B3 glenoids (−5.2 vs. −2.2 mm, p < 0.042). Changes in medial-lateral joint line (ΔML) were 2.5 ± 1.2 mm for B2 and 3.2 ± 1.4 mm for B3 glenoids, while CT-Vault measurements showed greater medialization in B3 glenoids (B2: 0.3 ± 1.6 mm; B3: −2.2 ± 2.1 mm; p < 0.001) (Table 4).14,17

Forward elevation improved in both B2 and B3 glenoids (preop: B2 112° ± 27.7, B3 100° ± 24.5, p = 0.2; postop: B2 156° ± 11.6, B3 150° ± 9.0, p = 0.103), with no significant differences between groups (Table 4).16 Joint line positioning improved in both subtypes (preop joint line relative to vault: B2 –2.2, B3 –5.2, p < 0.042, across A1, B2, B3), with correction toward premorbid anatomy greater in B2 than B3 glenoids (postop joint line relative to vault: B2 2.5° ± 1.2, B3 3.2° ± 1.4; CT-Vault: B2 0.3 ± 1.6, B3 –2.2 ± 2.1; p < 0.001) (Table 4).14 Humeral head centering improved in both groups (B2: −14% ± 7% preop to −1% ± 2% postop, p < 0.001; B3: −4% ± 6% preop to −1% ± 3% postop, P = 0.027), and post-operative radiolucency was observed in a subset of cases in both subtypes (B2: 31.4%; B3: 22.6%) (Table 4).12,13 Presence of central peg osteolysis at minimum of 2-year follow-up was greater in B3 glenoids managed with an augmented component compared with both B2 (augmented component) and A1 (standard component) (A1 SG: 2 [5%], B2 AG: 3 [10%], B3 AG: 6 [29%], P = 0.028).

3.6

3.6 Pain and patient satisfaction scores

For Constant Murley subdomains, B2 glenoids improved in pain (preop 6.4 ± 3.4, postop 13.4 ± 2.8, P = 0.001), ADL (preop 8.4 ± 3.6, postop 17.9 ± 3.6, P = 0.002), and strength (preop 1.7 ± 3.7, postop 11.5 ± 6.3, P = 0.043), while B3 glenoids improved in pain (preop 4.7 ± 2.7, postop 12.2 ± 3.2, P = 0.028) and ADL (preop 9.2 ± 3.7, postop 16.8 ± 2.8, P = 0.027) but not significantly in strength (preop 0 ± 0, postop 8.9 ± 5.7, P = 0.068) (Table 5).15 Postoperative VAS pain scores were similar between groups (B2: 1.3 ± 2.6; B3: 1.3 ± 2.8).12

Table 5 Pain and patient satisfaction scores.
Study VAS Pain Score Constant Murley Score Pain Constant Murley Score ADL Constant Murley Score Strength
Magosch 2021 N/A B2 preop: 6.4 ± 3.4 B2 preop: 8.4 ± 3.6 B2 preop: 1.7 ± 3.7
B2 postop: 13.4 ± 2.8 B2 postop: 17.9 ± 3.6 B2 postop: 11.5 ± 6.3
P = 0.001 P = 0.002 P = 0.043
B3 preop: 4.7 ± 2.7 B3 preop: 9.2 ± 3.7 B3 preop: 0 ± 0
B3 postop: 12.2 ± 3.2 B3 postop: 16.8 ± 2.8 B3 postop: 8.9 ± 5.7
P = 0.028 P = 0.027 P = 0.068
Polisetty 2023 B2 postop: 1.3 ± 2.6 N/A N/A N/A
B3 postop: 1.3 ± 2.8
4

4 Discussion

Historically, aTSA for GHOA with severe posterior glenoid bone loss was associated with high rates of instability due to a combination of edge loading, improper soft tissue balance, and inadequate fixation corridors.18–21 Previous studies have described a cascade of implant loosening involving decreased glenohumeral contact area, shear stress on the cement mantle, and glenoid component micromotion resulting in mechanical failure.22 However, recent studies have yielded different results and demonstrate durable long-term improvement in patients with B2 and B3 glenoids. This may be due to improved techniques, such as the use of eccentric reaming or augmented glenoid components, or could be related to the trend of treating more severe pathology with reverse total shoulder arthroplasty (rTSA).21 This study finds that patients with B2 and B3 glenoid morphologies see significant improvement in multiple clinical parameters after aTSA, a finding which challenges the assumption that these subtypes portend poor outcomes.21,23

In this review, we found that B2 glenoids exhibit significant improvements across all reported measures of range of motion after aTSA. In contrast, B3 glenoids showed improvements in flexion and abduction that did not reach statistical significance.15 The lack of significant improvement may be due to increased joint line medialization in B3 glenoids compared to B2 glenoids following aTSA.24 Biomechanical modeling studies have shown that joint line medialization leads to muscular shortening, increases deltoid muscle activity, and increased load on the supraspinatus tendon during arm elevation.24 However, the lack of significance should be interpreted cautiously, as available data were limited and sample sizes for B3 glenoids were small (n = 6). A different study reported greater improvement in internal rotation in B3 glenoids compared with B2 glenoids (Table 3).16 A recent study looking at rTSA demonstrated that increased post-operative glenoid baseplate retroversion was associated with improved internal rotation.25 Although performed in the context of rTSA, a similar biomechanical principal may apply to aTSA, in which B3 glenoids typically remain more retroverted than B2 glenoids, leading to increased internal rotation. Notably, the nonsignificant improvements in flexion and abduction were observed in aTSA with cementless convertible glenoid components while the relatively greater gain in internal rotation in the B3 group was achieved with a posteriorly stepped glenoid component, suggesting that implant design may play a role in the recovery of range of motion.15,16 Posteriorly stepped augmented components aim to restore the native joint line while minimizing glenoid bone stock loss from excessive anterior reaming.16 Compared to standard components, augmented components exhibit lower levels of joint line medialization, which has been associated with improved preservation of muscle function and joint stability.26 In this review, three studies used the Step Tech (DePuy Orthopaedics, Warsaw IN) posteriorly augmented glenoid, which is characterized by a dual-surface design with a posterior step of varying heights, allowing for correction of various degrees of retroversion.14,16,17,26 All three studies found significant improvement in PROMs and functional scores, with one study demonstrating greater improvements with augmented components compared with standard components.14,16,17

Both B2 and B3 glenoids demonstrated meaningful improvements in pain, Constant Murley, absolute, ADL, and strength scores. Both glenoid subtypes achieved substantial improvements in Constant-Murley strength although only the improvement B2 glenoids demonstrated reached statistical significance. B3 glenoids experienced a similar magnitude of improvement despite not reaching statistical significance, likely due to small sample size.15 These results are consistent with previous studies reporting significant pain relief in patients with posterior glenoid wear following aTSA.27 B3 glenoids exhibit a higher degree of retroversion than B2 glenoids and frequently require greater surgical correction of version, which can lead to joint medialization, decreased glenoid vault volume, decreased glenohumeral surface contact area, and increased contact pressures, possibly limiting pain relief.28

In terms of radiological findings, B2 and B3 glenoids demonstrated similar improvements in glenoid version, inclination, decentering, and HGA-AP. Despite being more retroverted preoperatively, B3 glenoids achieved greater overall correction and exhibited slightly better 2-year post-operative glenoid version compared to B2 glenoids.14 However, statistical significance cannot be established between B2 and B3 glenoids due to the inclusion of A1 glenoids in the comparison. Despite this, these results may reflect the effectiveness of aTSA in correcting severe deformities. B3 glenoids demonstrated joint lines that remained persistently medialized relative to premorbid anatomy post-operatively compared to B2 glenoids.14 According to Ianotti et al. this difference is likely due to the need for more anterior glenoid reaming when using a posteriorly stepped augmented glenoid component in more severe B3 glenoids.14 These findings suggest that severe posterior wear may limit full restoration of the premorbid joint line, even with the use of a stepped augmented glenoid component. Richetti et al. further suggested a thicker glenoid component or a wedge-shaped augmented component may better match the bone deformity of a B3 glenoid, whereas posteriorly stepped components more closely match the B2 deformity.17 Despite having more posterior bone loss, B3 glenoids were found to exhibit lower rates of post-operative radiolucency than B2 glenoids (Table 4).12 This finding suggests that standard glenoid components may provide stable fixation even in advanced deformities.12 Interestingly, when treated with augmented components, B3 glenoids demonstrated a higher presence of central peg osteolysis compared to B2 (treated with augmented component) and A1 (treated with standard component).14 This difference is likely associated with greater levels of medial bone loss in B3 glenoids pre-operatively, resulting in additional anterior reaming required when using a posteriorly stepped augmented component, which may increase stress on the central peg.14 Finally, while the differences in HSA-AP were significantly different between A1, B2, and B3 glenoids pre- and post-operatively, HSA-AP has been found to be correlated to glenoid version.14,17,29 Therefore, these findings were likely influenced by differences in glenoid component version.14,17

This study has several limitations. The heterogeneity of study design and measured outcomes could introduce variability and limit generalizability. In addition, differences in surgical techniques used amongst included studies may alter the true differences in outcomes between the two glenoid subtypes. Furthermore, the small number of B3 glenoids in many studies may limit statistical power and underestimate differences in outcomes. Finally, confounding factors such as age, sex, and comorbidities may influence functional outcomes and patient satisfaction.

5

5 Conclusion

This review demonstrates that aTSA provides meaningful improvements in joint function, pain, range of motion, strength, and patient satisfaction for B2 and B3 glenoids. Augmented glenoid components can effectively address the challenges presented by severe posterior wear. While improvements were observed in both glenoid subtypes, direct comparisons between B2 and B3 outcomes were limited. Additionally, several outcomes in B3 cohorts did not reach statistical significance, and sample sizes for these groups were consistently small. As a result, the available evidence should be interpreted as descriptive rather than definitive regarding comparative outcomes between deformity types. Future studies with larger B3 cohorts and longer follow-up time are needed to better understand optimal surgical treatment and long-term outcomes.

Ethical approval

Not applicable.

Contributorship

ZW: Conceptualization, formal analysis, investigation, writing - original draft, writing - review and editing. AKC: Formal analysis, writing - original draft. SW: Conceptualization, supervision, project administration, writing - review and editing. MSS: Writing - review and editing. ASC: Writing - review and editing. JS: Writing - review and editing. JC: Resources. JV: Writing - review and editing.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

  1. , , . Optimal management of glenohumeral osteoarthritis. Orthop Res Rev. 2018;10:9-18.
    [Google Scholar]
  2. , , , , . Glenohumeral osteoarthritis: an overview of etiology and diagnostics. Scand J Surg. 2021 Sept;110(3):441-451.
    [Google Scholar]
  3. , , , , . Reverse and anatomic shoulder arthroplasty regional usage and open payment analysis using the centers for medicare and medicaid services database. J Shoulder Elb Arthroplast. 2024;8
    [Google Scholar]
  4. , , , , , , . Utilization of shoulder arthroplasty in the United States – an analysis of current trends and future predictions. Semin Arthroplasty JSES. 2020 Sept 1;30(3):200-209.
    [Google Scholar]
  5. , , , et al . Anatomic versus reverse shoulder replacement: are we asking the right questions and what are the answers? J Hand Microsurg. 2025 May 1;17(3):100225.
    [Google Scholar]
  6. , , , , , . Glenohumeral osteoarthritis and reverse shoulder replacement. JSES Int [Internet] 2024 July 5
    [Google Scholar]
  7. , , , . When should reverse total shoulder arthroplasty be considered in glenohumeral joint arthritis? Clin Shoulder Elb. 2021 Dec;24(4):272-278.
    [Google Scholar]
  8. , , , , . Classifications in brief: Walch classification of primary glenohumeral osteoarthritis. Clin Orthop. 2017 Sept;475(9):2335-2340.
    [Google Scholar]
  9. , , , , . Evolution of the walch classification and its importance on the B2 glenoid. J Shoulder Elb Arthroplast. 2020 Jan;4
    [Google Scholar]
  10. , , , , , , . A modification to the walch classification of the glenoid in primary glenohumeral osteoarthritis using three-dimensional imaging. J Shoulder Elb Surg. 2016 Oct;25(10):1601-1606.
    [Google Scholar]
  11. , , , , . Quantitative measurement of osseous pathology in advanced glenohumeral osteoarthritis. J Bone Joint Surg Am. 2017 Sept 6;99(17):1460-1468.
    [Google Scholar]
  12. , , , et al . Anatomic and reverse shoulder arthroplasty for management of type B2 and B3 glenoids: a matched-cohort analysis. J Shoulder Elb Surg. 2023 Aug;32(8):1629-1637.
    [Google Scholar]
  13. , , , , . Anatomic total shoulder arthroplasty with all-polyethylene glenoid component for primary osteoarthritis with glenoid deficiencies. JB JS Open Access. 2020;5(4)
    [Google Scholar]
  14. , , , , . Stepped augmented glenoid component in anatomic total shoulder arthroplasty for B2 and B3 glenoid pathology: a study of early outcomes. J Bone Jt Surg. 2021 Oct 6;103(19):1798-1806.
    [Google Scholar]
  15. , , , , , . Prospective midterm results of a new convertible glenoid component in anatomic shoulder arthroplasty: a cohort study. Arch Orthop Trauma Surg. 2021 May;141(5):717-724.
    [Google Scholar]
  16. , , , et al . Factors associated with functional improvement after posteriorly augmented total shoulder arthroplasty. J Shoulder Elb Surg. 2023 June;32(6):1231-1241.
    [Google Scholar]
  17. , , , et al . Three-dimensional computed tomography analysis of pathologic correction in total shoulder arthroplasty based on severity of preoperative pathology. J Shoulder Elb Surg. 2021 Feb;30(2):237-249.
    [Google Scholar]
  18. , , . Influence of preoperative factors on outcome of shoulder arthroplasty for glenohumeral osteoarthritis. J Bone Joint Surg Am. 2003 Feb;85(2):251-258.
    [Google Scholar]
  19. , , , , . Predictors of medium-term clinical outcomes after total shoulder arthroplasty. Arch Orthop Trauma Surg. 2017 Feb;137(2):187-193.
    [Google Scholar]
  20. , , . Approach to glenoid bone loss and deformity in B3 and C glenoids: primary anatomic shoulder arthroplasty. Semin Arthroplasty JSES. 2024 Mar;34(1):259-265.
    [Google Scholar]
  21. , , , et al . Short-term outcomes of anatomic total shoulder arthroplasty with nonaugmented glenoid component for Walch B2 and B3 glenoid morphology. J Shoulder Elb Surg. 2025 June;34(6):S10-S21.
    [Google Scholar]
  22. , , , , , , . Posterior glenoid wear in total shoulder arthroplasty: eccentric anterior reaming is superior to posterior augment. Clin Orthop. 2015 Dec;473(12):3928-3936.
    [Google Scholar]
  23. , , , et al . Mid- to long-term outcomes of augmented and nonaugmented anatomic shoulder arthroplasty in walch B3 glenoids. J Shoulder Elb Surg. 2022 June;31(6S):S103-S109.
    [Google Scholar]
  24. , , , , , , . Joint-line medialization after anatomical total shoulder replacement requires more rotator cuff activity to preserve joint stability. JSES Int. 2021 May;5(3):406-412.
    [Google Scholar]
  25. , , , et al . Increased glenoid baseplate retroversion improves internal rotation following reverse shoulder arthroplasty. JSES Int. 2025 Jan;9(1):147-154.
    [Google Scholar]
  26. , , . Anatomic augmented glenoid implants for the management of the B2 glenoid. J Shoulder Elb Arthroplast. 2019;3
    [Google Scholar]
  27. , , . Decision making in glenohumeral arthroplasty. J Arthroplast. 2003 Jan;18(1):75-82.
    [Google Scholar]
  28. , , , , , . Structural bone grafting for glenoid bone loss in primary anatomic total shoulder arthroplasty: a case series and technique report. JSES Rev Rep Tech. 2024 Aug;4(3):464-471.
    [Google Scholar]
  29. , , , , . 3D CT assessment of the relationship between humeral head alignment and glenoid retroversion in glenohumeral osteoarthritis. J Bone Jt Surg. 2014 Apr 16;96(8)
    [Google Scholar]

Fulltext Views
39

PDF downloads
6
View/Download PDF
Download Citations
BibTeX
RIS
Show Sections

Suggested read for related articles: