2.1.3. Convertible Platforms

With the dramatic rise of primary shoulder arthroplasty over the last 15–20 years has come a rise of revision shoulder arthroplasty. Though indications for revision shoulder arthroplasty vary, it remains a technically demanding and challenging procedure regardless of indication. Revision shoulder arthroplasty is associated with increased blood loss and operative times, and frequently requires the use of special implants, augments, and bone grafting [26]. The innovative development of modular platform humeral stems, those

that can be converted from an anatomic TSA construct to a reverse TSA, has significantly reduced the complexity of revision shoulder arthroplasty. It theoretically obviates the need to explant the humeral stem that in turn reduces operative time, decreases blood loss, preserves humeral bone stock, and can reduce cost [27]. Crosby et al. found a slightly better postoperative range of motion in those who underwent conversion to RSA with a convertible platform compared to those that required entire humeral implant exchange. The prevalence of intraoperative complications was significantly lower with the convertibleplatform group (0% compared to 15%), though rates of reoperation were not different [27].

**Figure 2.** Short stem anatomic TSA. AP X-ray of anatomic TSA with short humeral stem.

It should be noted that not all convertible-platform humeral components can indeed be retained at the revision setting. The convertible platform humeral stem must be well-fixed and well-positioned to be retained. Encouragingly, multiple studies have demonstrated that a vast majority, approximately 80%, can be retained [27–29].

#### *2.2. Glenoid Component Innovations*

Augmented Glenoid Components

Excess retroversion and posterior wear of the glenoid present a dilemma for the orthopedic surgeon performing anatomic TSA. Glenoids with posterior wear and formation of a neo-glenoid (Walch B2), and those with >15◦ retroversion (Walch B3) are at high risk of developing glenoid loosening with standard implants [30]. Glenoid loosening has been the most common cause of anatomic TSA failure and indication for revision. Increased osteolysis has been demonstrated in cases in which the glenoid component is placed in excess retroversion, resulting in decreased implant survival [31].

Glenoid component retroversion beyond 15◦ leads to decreased contact area and increased contact pressures, placing the glenoid component at high risk of failure [31], and though eccentric reaming can correct small posterior deficits up to 10–15◦ [32,33], one risks removing excessive native bone when eccentrically reaming for larger deficits. Primary bone grafting has demonstrated variable results and may be associated with clinical and radiographic failure [34–37]. The other remaining option to deal with excessive posterior wear and retroversion even in the setting of an intact rotator cuff is to perform RSA, and some authors prefer this method [30].

For these reasons, there has been the introduction of augmented glenoid components that theoretically reduce bone removal and shear stresses, while retaining the benefits of anatomic TSA over RSA. Full wedge and partial wedge augments of varying degrees exist. Older designs included a keel that was angled in line with the neo-glenoid face, thus directing fixation toward the anterior neck of the glenoid. Newer implants have placed vault fixation angled with the paleo-glenoid to improve fixation [38]. Strong long-term data is lacking for augmented glenoid components. However, the short-term results are encouraging with multiple studies citing revision rates of 0%–5% at 2–3-year follow-up [38–43]. Larger-degree augments may be at higher risk of failure, as demonstrated by Priddy et al. in their retrospective study of full-wedge glenoid augmented TSA compared to nonaugmented TSA, in which all failures of the augmented glenoids requiring revision came with the 16◦ augment, with no failures of the 8◦ or 12◦ augments. There were no differences in radiographic lucencies around the pegs, postop ROM or patient reported outcome measures [38].

The senior author's preference for managing glenoid retroversion includes high-side reaming a B2 or B3 glenoid for retroversion <25◦. When retroversion is 25◦ to 35◦, the preference is to use an augmented glenoid component (Figure 3). For retroversion >35◦, the senior author will perform an RSA with bone grafting of the glenoid.

**Figure 3.** Anatomic TSA with posteriorly augmented glenoid polyethylene. Axillary X-ray of posteriorly augmented glenoid polyethylene.

#### *2.3. Inlay versus Onlay Glenoid Components*

Traditional onlay glenoid prostheses exhibit signs of loosening at relatively high rates, even when optimally placed [44–46]. Metal-backed glenoids have fallen out of favor due to the unacceptably high failure rates [47], so all-polyethylene designs are the gold standard. Though somewhat controversial, pegged onlay glenoids appear to have superior survivorship to keeled glenoids [45,48]. The "rocking-horse" phenomenon at the glenoidbone interface can result in edge loading, liftoff, and subsequent component loosening. With radiolucent lines occurring in approximately 30%–75% of TSAs with onlay glenoids by 10-year follow up, and loosening resulting in clinical failure requiring revision TSA 2%–10% of the time [49–51], there indeed is need for improvement in implant design. This has led to the development of the inlay glenoid design, in which the polyethylene component is implanted flush with the glenoid bone surface. Its theoretical advantages are those of less glenoid bone removal and improved mechanical characteristics due to less implant edge loading and lift off.

In a cadaveric study by Gagliono et al., onlay glenoid components exhibited gross loosening during fatigue testing, whereas the inlay glenoid components did not, and the onlay glenoids experienced significantly higher forces acting on them than did the native or inlay glenoids [52]. Short term results are promising, with good improvement in PROs, function, and ROM, without increased complication rates, and low reoperation rates [53]. This has been true even in the setting of posterior glenoid erosion, with no differences in short term clinical and radiographic outcomes evaluating non-spherical humeral head and inlay glenoid components in concentric (Walch A) glenoids compared to non-concentric (Walch B1 and B2) glenoids, according to the work of Egger et al. [54]. Inlay components may be of particular benefit in the younger, athletic, weight-lifting population given the theoretical decrease in mechanical loosening and resultant lack of restrictions afforded to them. Early clinical results have been excellent, and most of these patients are able to return to sports and lifting at the same or higher level [53,55].

Longer term data is required to definitively say whether inlay glenoid components are superior to onlay components, but early evidence suggests that this may turn out to be the case.

The senior author typically uses an onlay glenoid component, except when glenoid dysplasia exists, in which case the preference is to use an inlay glenoid component. See Figure 4, which depicts an X-ray appearance of an inlay glenoid component.

**Figure 4.** Inlay glenoid polyethylene in setting of glenoid dysplasia. Axillary X-ray of inlay glenoid polyethylene in setting of glenoid dysplasia.

#### *2.4. Convertible Glenoid Components in Anatomic TSA*

Cemented all-polyethylene glenoid components have represented the gold standard in anatomic TSA, given the historically unreliable results of cementless glenoid components. However, given the challenges and risks of revising a cemented glenoid component, there has been a resurgence of interest in convertible metal-backed glenoid components for anatomic TSA. The new generation of convertible metal-backed trays feature improved designs including a highly stable anchorage mechanism of the metal carrier in the glenoid vault, with larger bone-implant contact area and improved stability against shear forces [56]. Short and midterm follow-up results of the latest generation of convertible glenoid systems are encouraging, with revision rates ranging from 0%–11% [56–58]. Magosch et al. reported no glenoid loosening, an implant related revision rate of 4.2%, polyethylene dissociated in 4.2%, and no complications in cases requiring revision to RSA, in their prospective study of 48 patients at a mean follow up of 49 months [56]. In the setting of failed anatomic TSA, conversion to RSA may be facilitated by convertible glenoid systems, while maintaining improvements in pain and shoulder function [59]. Long-term follow-up data is needed, but there may indeed be a role for these convertible glenoid components moving forward.

#### **3. New Perspectives and Innovations in Reverse Shoulder Arthroplasty**

Reverse shoulder arthroplasty was approved by the FDA in the US in 2003. Since then, the prevalence has increased significantly by more than 2.5 times from 7.3 cases per 100,000 persons to 19.3 cases per 100,000 persons between 2012 to 2017 [60]. Its original indication was rotator cuff arthropathy in older patients [61], but this has since been expanded as prostheses have improved and surgeon experience has become more ubiquitous. Indications now include fracture, revision shoulder surgery, rotator cuff arthropathy in relatively younger patients, tumor, and glenoid bone loss. As these indications expand, more options have developed to assist in decreasing complications and improving complex or revision surgeries.

#### *3.1. Combined Humeral and Glenoid Component Innovations* Lateralization

Reverse shoulder arthroplasty designs have evolved over the last 15 years. One major evolution has been increased lateralization of the glenoid component. Typically, in RSA, the glenoid center of rotation (COR) is still medialized relative to the native shoulder COR. The increased lateralization is in reference to the preoperative humeral position rather than the COR. The current trend is to increase lateralization to increase soft tissue tension, particularly the rotator cuff. This lateralization can be achieved by one of three methods: lateralized glenoid baseplate, lateralized glenosphere or glenoid bone grafts (e.g., BIORSA).

Lateralization can have negative and positive effects on both the glenoid and humerus. Glenoid lateralization decreases adduction impingement thereby decreasing scapular notching and improving adduction, ER and extension motion [62–66]. It also improves rotator cuff tension and prosthetic stability [67,68]. Glenoid lateralization does, however, decrease the mechanical advantage of the deltoid, increase the shear forces across the implant interface and increase acromial strain, potentially increasing the risk of stress fracture [69–74]. Humeral lateralization on the other hand improves the deltoid mechanical advantage as well as improves the posterior cuff tension and the deltoid wrap by providing a more anatomic vector of muscle pull [67,70,75,76]. The negative effects of humeral lateralization are potentially too much soft tissue tension when combined with glenoid lateralization.

#### *3.2. Humeral Component Innovations*
