*1.3. Reverse Arthroplasty*

Grammont revolutionized the field of shoulder arthroplasty with the development of the reverse prosthesis concept: a more constrained implant of reverse geometry that would increase the moment arm of the deltoid to compensate for the rotator cuff insufficiency [8]. Despite becoming an incredibly successful implant in terms of the restoration of active elevation and the long-term survivorship, [9] the limitations of the traditional Grammontstyle prosthesis included poor restoration of the internal and external rotation as well as excessive impingement of the humeral polyethylene with the medial scapular pillar, leading to polyethylene wear, bone loss (notching), and eventually loosening [10]. Frankle modified the reverse principles to optimize the impingement-free range of motion and the tension of the axial rotator cuff with the design of a reverse prosthesis with a less truncated sphere and a more vertical polyethylene (135-degree opening angle) [11,12]. Contemporary designs follow the modifications of reverse introduced by Frankle.

#### *1.4. Shorter Stems, Resurfacing, and Stemless*

The length of most of the traditional stems was arbitrarily set to occupy the upper third to half of the humerus. Standard-length stems have demonstrated outstanding performance and survivorship [13]. However, avoiding relatively long stems is attractive for several reasons, including easier revision, easier implantation of an ipsilateral total elbow arthroplasty, and maybe easier managemen<sup>t</sup> of periprosthetic fractures. Resurfacing arthroplasty represented a first attempt to avoid stem use, [14] but resurfacing components have fallen out of favor since incomplete head removal made glenoid access more difficult, and the prosthetic head sizes with various degrees of offset and eccentricity could not be used. As such, the design has evolved into the use of short-stem and stemless prostheses. The length of most of the short stems has been chosen arbitrarily, with few exceptions. For anatomic arthroplasty, stemmed and stemless prostheses seem to provide equivalent results, provided satisfactory implantation is achieved at the time of surgery [15,16].

#### *1.5. Malalignment and Adverse Bone Reactions*

One benefit of longer stems is that a tight stem fit into the endosteal canal facilitates adequate alignment. Ultrashort stems introduce two potential problems: poor alignment and adverse bone reactions.

Stems that do not engage the cylindrical portion of the endosteal canal can easily be misaligned. Excessive valgus or varus will lead to a poor humeral head position in anatomic arthroplasty. Similarly, poor alignment can lead to a reverse polyethylene that is excessively horizontal with an increased risk of notching, or to a more vertical polyethylene that may facilitate dislocation. Certain short stems have been designed with just enough length to avoid malalignment [17]. Stemless prostheses are also at risk for malalignment (Figure 1). As such, care must be taken to optimize the humeral head cut to minimize the chances of malalignment with ultrashort stem and stemless prostheses.

Certain ultrashort stems need larger diameters to achieve primary stability in the absence of diaphyseal contact. This concept has been captured with the fill–fit ratio popularized by Walch et al. [18]. Severe stress shielding with resorption of the greater tuberosity, and in extreme cases in areas of complete cortical defect, has been reported with the implantation of larger sizes of certain stems, and malalignment may accentuate these adverse bone reactions through point contact of the stem on the cortical bone, further shielding the proximal bone from stress (Figure 2). Thus, it is important to design implants that do not shield the metaphysis from stress.

**Figure 1.** Ultrashort stems (**A**) and stemless prostheses (**B**) are at increased risk for malalignment.

**Figure 2.** Certain ultrashort stems are associated with substantial stress shielding.

#### *1.6. Preoperative Planning Software and Surgical Execution*

The development and widespread use of preoperative planning software has revolutionized the field of shoulder arthroplasty. I trained at a time when plain radiographs were the only imaging study obtained before shoulder arthroplasty. Today, the vast majority of shoulder arthroplasty surgeons rely on computer tomography to understand each shoulder to be replaced and to plan the surgery accordingly. Furthermore, preoperative planning software has advanced the field to a whole other level: three-dimensional renderings,

automated measurements, and virtual implant overlays allow for accurate planning of the implant positioning to optimize orientation, seating, contact, motion free of impingement, and other variables [19–21]. Such software can then be used for artificial intelligence predictive algorithms, manufacturing patient-specific guides, and using computer-assisted surgery with navigation or robotics.

In the field of reverse shoulder arthroplasty, preoperative planning software reveals that using a larger glenoid with a larger lateral offset and an inferior overhang is the most successful strategy to optimize the range of motion free of impingement, especially when combined with a more vertical (typically 135 degree) polyethylene opening angle [22].

#### *1.7. Same-Day Surgery and Ambulatory Surgery Centers*

In the United States, there is a growing interest in same-day discharge after shoulder arthroplasty, as well as in performing these procedures in ambulatory surgery centers. This is driven by two main forces: the potential for certain financial gain and the need to decrease hospitalizations, especially considering the current COVID-19 pandemic [23]. Ambulatory surgery centers have less capacity to process large inventories and instrument trays. As such, there is the need for streamlined instrumentation and shared instruments between stem and stemless designs. Patient-matched implants and preoperative planning software may further help decrease inventory.

#### *1.8. Proximal Humerus Bone Density*

Understanding the bone mineral density of the proximal humerus is paramount to optimizing the primary stability of modern humeral components. In the osteoarthritic shoulder, the strongest bone is at the periphery and is closer to the superior aspect of the humeral head. As such, fixation is theoretically optimized by achieving a prosthetic fit to the periphery of the metaphysis and with a slightly higher humeral head cut [24]. However, one downside of performing a higher humeral head cut is the more difficult access to the glenoid.

#### *1.9. Implications for Humeral Component Design*

The brief historical review summarized above provides the grounds for design features that are perceived as desirable when considering contemporary humeral component design (Table 1).

**Table 1.** Desirable features for contemporary humeral component design.

	- - Preoperative planning software
	- - Cutting guides
	- - Patient-specific guides
	- - Navigation
	- - Robotics

#### **2. Implant Configurations: What Are Our Targets on the Humeral Side Currently?**

*2.1. Anatomic Shoulder Arthroplasty*

When performing an anatomic shoulder arthroplasty, the main goal on the humeral side is to restore the overall geometry of the proximal humerus. Considering the variability of the human shoulder (in terms of size, retroversion, and other parameters), as well as the need to adapt to the final position of the humeral stem or stemless nucleus, it is necessary to design systems with multiple head sizes and thicknesses as well as with a mechanism to offset the humeral head with respect to the final position of the humeral stem/nucleus (Figure 3). In most shoulders, the restoration of the premorbid anatomy provides the best outcome. However, in certain shoulders, the humeral head version, diameter, thickness, and/or eccentricity may need to be adapted to the condition of the soft tissues. For example, in a shoulder with substantial posterior subluxation and chronic stretching of the posterior rotator cuff and capsule, it may be necessary to implant a humeral head that is thicker than the premorbid native head to properly tension the soft tissue envelope posteriorly.

**Figure 3.** Options for replacement of the humeral head with one system for anatomic shoulder arthroplasty.

#### *2.2. Reverse Shoulder Arthroplasty*

Understanding the nuances associated with the design and implantation of the humeral component in reverse shoulder arthroplasty is not possible without considering the glenoid side [22]. Currently, most would agree that reverse shoulder arthroplasty requires a fine balance between (1) maximizing impingement-free range of motion and (2) optimizing soft tissue tension and muscle function around the shoulder.

Avoiding any impingement between the medial aspect of the polyethylene and the body of the scapula and scapular pillar essentially requires displacing the proximal humerus laterally and posteroinferiorly. This is best achieved by implanting a larger glenosphere with a posteroinferior overhang in reference to the glenoid vault combined with a larger lateral offset of the glenoid component. Larger lateral offsets may be achieved with thicker glenospheres, structural bone grating between the native glenoid and the baseplate (bio-RSA), or thicker (augmented) baseplates (Figure 4). The benefit of bio-RSA and augmented baseplates over thicker glenospheres is that both bone graft and metal augments provide adequate correction of angular deformities (inclination and retroversion) without reaming excessively, which can lead to impingement as well.

**Figure 4.** (**A**), Impingement-free range of motion is optimized with implantation of larger glenospheres with a lateral offset and an inferior overhang. Glenoid lateralization may be achieved with thicker glenospheres (**B**), the use of a bone graft under the baseplate (BIO-RSA), or the use of augmented baseplates (**C**).

Provided the surgeon commits to implantation of large glenospheres with an inferior overhang and a lateral offset using any of the three methods above, the humeral component must allow for a minimal thickness above the cut surface in order to avoid excessive soft tissue tension secondary to lateralization, distalization, or both. A relatively easy way to design humeral components that allow for anatomic and reverse compatibility is to design humeral bearings that rest on the cut surface of the humerus, so-called onlay systems. The downside of onlay systems for those surgeons who maximize impingement-free motion on the glenoid side is that the soft tissue tension may be excessive. This can be compensated for by lowering the humeral cut, which may be acceptable in the cuff-deficient shoulder but not in the cuff-intact osteoarthritic shoulder, where a lower cut would damage the rotator cuff. As such, if the surgeon chooses to maximize impingement-free range of motion through glenoid implantation, the thinnest humeral bearing construct should place the pivot point at or below the humeral cut. When the pivot point (the deepest portion of the polyethylene) is below the humeral cut, implants are classified as inlays.

However, *the onlay vs. inlay controversy should probably be abandoned* for two reasons. Firstly, there is a high level of variability regarding how much lateralization and distalization are provided by the many implants in the market [25]. Classifying them as onlays or inlays is an oversimplification. What matters is where the humerus "lands" for a specific glenoid reconstruction, depending on the humeral implant selected and where it is implanted. This will affect the length and the moment arm of the deltoid and rotator cuffs [26]. Secondly, surgeons may implant inlay components in an onlay fashion or the other way around. For example, the original prosthesis designed by Dr. Frankle was an inlay design; however, its proximal portion was relatively large and could not be fully inset in the humeral metaphysis of many patients, thus resulting in an onlay application of an inlay design (Figure 5). By the same token, if thicker polyethylene bearings or a metal spacer are added to an inlay prosthesis to guarantee adequate stability, the pivot point is at an onlay level despite the implant being designed as an inlay. Consequently, even though implants that allow placement of the pivot point at or below the cut surface of the humerus are necessary to optimize soft tissue tension across the whole spectrum of shoulder replacements, in many shoulders, these inlay components will behave as onlay ones because thicker polyethylenes may be needed to avoid dislocation, especially in the cuff-deficient shoulder. The ideal degree of humeral lateralization probably varies from individual to individual depending on the underlying diagnosis and other characteristics.

**Figure 5.** This implant was designed as an inlay, but its large size resulted in an onlay application most of the time.

#### **3. From Design to Implantation: Pearls and Pitfalls Learned with Use of a Contemporary Humeral Component**

Hopefully, a review of the history of implant component design and an understanding of what are considered contemporary targets today will help drive the surgical techniques for implantation of contemporary humeral components (Figure 6).
