The glenohumeral joint is a complex, highly mobile joint that relies on the bony articulation between the humeral head and glenoid as well as the surrounding soft tissues for stability and balance.¹ The goal of shoulder arthroplasty is to re-create normal glenohumeral anatomy.² This can be a challenge not only because pathologic changes to the glenoid and humeral head can distort normal anatomy but also because of the inherent variation in “normal” anatomy that exists among patients.¹,³,⁴ When “normal” glenohumeral anatomy is more closely restored, shoulder arthroplasty results in better functional outcomes and longer implant survival.¹,⁵,⁶,⁷ This led to the development of glenoid and humeral implants that can accommodate the variations encountered. Furthermore, as our understanding of the anatomy and biomechanics of the glenohumeral joint has expanded, we have an opportunity to develop improved methods of preoperative planning and intraoperative navigation that could, ultimately, result in improved outcomes. In this chapter, we will review glenohumeral anatomy and its relationship to implant design.

The glenoid height, width, surface area, version, inclination, vault shape and size, and radius of curvature are all important anatomic parameters for prosthesis design. Multiple cadaveric and patient studies have reported the differences in these anatomic parameters among patients.

Glenoid height is the measurement from the most superior and to the most inferior portion of the glenoid (FIGURE 2.1; TABLE 2.1). Checroun et al evaluated 412 cadaver scapulae using direct measurement. These scapulae had an average age of 58 years (range, 24-87 years). The average glenoid height was 37.9 mm (range, 31.2-50.1 mm).⁹ Kwon et al evaluated 12 cadaveric scapulae by CT imaging. The average glenoid height was 39.1 mm (range, 31-48 mm).¹² McPherson et al assessed 93 cadaveric shoulders radiographically using a custom computer software program. They reported an average glenoid height to be 33.9 mm.¹⁹ Gender differences in glenoid height was assessed by Churchill et al in 344 scapulae with an average age of 25 years (range, 20-30 years) using direct measurement. The average glenoid height for male and female specimens was 37.5 and 32.6 mm, respectively.¹⁰ Lastly, Moineau et al assessed 41 arthritic glenoids using CT images. The average glenoid height was 41.33 mm (range, 31.5-55.1 mm), which supports the observation that glenohumeral arthritis results in glenoid enlargement primarily as a result of adaptive changes and osteophyte formation.⁴

Glenoid width is the measurement from the anterior rim to the posterior rim (FIGURE 2.1; TABLE 2.1). However, glenoid width varies depending on the level where the measurement is obtained because of the shape of the glenoid. Based upon anatomic studies, the glenoid shape or face is described as elliptical or pear shaped, with pear shaped being described in 71% of the specimens9 (FIGURE 2.1). Checroun et al reported the average glenoid width at the midportion of the 412 cadaveric scapulae to be 29.3 mm (range, 22.6-41.5 mm).⁹ Kwon et al reported the average width to be 25.2 mm (range, 21-34 mm), while McPherson et al described the average width to be 28.6 mm.¹²,¹⁹ Churchill et al compared glenoid width between genders. From their review of 344 scapulae, the average glenoid width for male and female specimens was 27.8 and 23.6 mm, respectively.¹⁰ Lastly, Moineau described the average glenoid width to be 29.35 mm (range, 19.6-44.8 mm) in 41 arthritic shoulders, again documenting the relative enlargement of the arthritic glenoid.

The glenoid surface area is the area of the entire articular surface (FIGURE 2.1). The variation in glenoid height and width results in a similar variation in glenoid surface area. Kwon reported the average glenoid surface area in 12 cadaveric scapulae based upon CT imaging to be 8.7 cm² (range, 7.0-14.2 cm²).¹²

Glenoid version is the angle of the glenoid face with respect to the transverse axis of the scapula (FIGURE 2.2; TABLE 2.1). Anteversion is present when the glenoid face is angled anteriorly with respect to the transverse axis, and retroversion is present when the glenoid face is angled posteriorly. Measuring glenoid version has been described by different methods. However, glenoid version is most often calculated using Friedman’s method.²,¹¹,²⁰,²¹,²² Based on aggregated studies reporting glenoid version in
1234 specimens, the average glenoid version was 6.3° of retroversion (range, 1°-12.1° retroversion).²,⁴,⁹,¹⁰,¹¹,²⁰,²¹,²²,²³,²⁴ Even though the range of the combined averages for these studies was 1° to 12.1° of glenoid retroversion, individual studies have reported glenoid version to be up to 23.2° of anteversion and up to 32° of retroversion, which clearly shows the wide variation of anatomic version in nonarthritic glenoids.⁴,²⁰

Glenoid inclination is the slope of the articular surface from superior to inferior and is also based upon the transverse axis as a reference (FIGURE 2.3). Superior inclination is present when the glenoid articular surface is facing superiorly; conversely, with inferior inclination, the glenoid articular surface faces inferiorly. Ricchetti evaluated 57 shoulders with CT imaging and reported an average of 10° superior inclination.² Boileau assessed 47 shoulders with radiographs and CT images and reported glenoid inclination to be 15° superior based upon CT imaging and 10° superior based upon standard radiographs.²⁵ Gender differences were evaluated by Churchill who reported glenoid inclination to be 4° superior in men (range, 7° inferior to 15.8° superior) and 4.5° superior in women (range, 1.5° inferior to 15.3° superior).¹⁰ Of note, this study emphasized the wide variability of inclination values encountered in the 344 cadaveric scapulae studied.

Various methods have been described to measure the glenoid inclination. At present, the beta angle is considered the most accurate method.²,¹⁰,²⁵,²⁶ The beta angle is measured based upon the intersection of a line drawn on the floor of the supraspinatus fossa and a line drawn on the glenoid fossa that connects the superior and inferior margins²⁶ (FIGURE 2.4). Measuring glenoid inclination using the beta angle has been shown to be more accurate using CT coronal images than standard radiographs.²⁶ The beta angle has also been referred to as the total shoulder arthroplasty angle and the global glenoid inclination.²⁵ The beta angle is referred to as the global glenoid inclination as it incorporates the entire glenoid surface when measuring inclination. Typically, glenoid components for anatomic shoulder replacements are
implanted covering the entire glenoid; therefore, using a measurement that takes the entire glenoid into account is reasonable. In contrast, reverse shoulder arthroplasty baseplate designs utilize smaller baseplates that only cover the inferior portion of the glenoid. Therefore, an inclination measurement focusing only on the inferior aspect of the glenoid has been proposed for inclination measurements for smaller baseplates that, when implanted, cover only the inferior aspect of the glenoid.²⁵ The reverse shoulder arthroplasty angle obtains an inclination measurement on the more inferior aspect of the glenoid surface for rotator cuff-deficient shoulders that will require a reverse shoulder prosthesis.²⁵