Glenoid baseplate position in reverse shoulder arthroplasty

Abstract

Background

With the growing popularity of reverse shoulder arthroplasty and rapid advancements, attention has returned to the role of baseplate positioning on outcomes and complications. The purpose of this review is thus to revisit the current state of literature as it relates to glenoid baseplate positioning from both the clinical and biomechanical perspective.

Methods

An extensive literature search was conducted to review relevant biomechanical and clinical studies investigating outcomes of variations in baseplate positioning and inclination and discuss how these parameters may be changing with modern implant designs.

Results

Glenoid baseplate positioning continues to evolve as we have yet to establish a clear consensus on the gold standard of placement. Inferior placement of the glenoid may not be necessary based on the growing popularity of lateralized glenoid design. Inferior glenoid tilt is biomechanically stronger; however, the use of a lateralized glenoid may negate the negative effects previously seen with superior glenoid tilt.

Conclusion

There remains little knowledge regarding the optimal glenoid version. Given the rapid design developments of the reverse shoulder arthroplasty, the importance of glenoid baseplate positioning on functional outcomes may need to be revisited.

Management of the rotator-cuff deficient arthritic shoulder has presented longstanding challenges to the orthopedic surgeon. Early unconstrained shoulder arthroplasty systems (ie, the Bickel prosthesis, Stanmore prosthesis) were associated with high complication and implant failure rates^20,35 Given these difficulties, the reverse shoulder arthroplasty (rTSA) explored a reverse ball-and-socket design with hopes of addressing scapular fixation and improving functional outcomes.

In 1985, Paul Grammont created a novel system focusing on four key principles: inherent stability of the prosthesis, convexity of the weight bearing portion, medialization and distalization of the center of rotation to increase deltoid function, and that the sphere's center must be within the glenoid neck.²⁴ While the Grammont prosthesis has revolutionized rTSA, it is not without complications. The torsional forces and constraints created by the implant can lead to humeral changes such as osteolysis, osteopenia, and medial and lateral cortical bone narrowing.^5,37 In addition, complications inherent to rTSA including scapular notching, acromial stress fractures, and instability remain present.^1,19,53

In recent years, there has also been a shift away from the Grammont design, namely to glenoid-sided lateralization and humeral sided changes. The Grammont prosthesis incorporates a medial glenosphere with a 155° humerus. Now, most implants have a lateralized glenoid and more varus humeral neck shaft angles.⁴² In addition, glenoid lateralization had an influence on scapular notching. While glenoid lateralization is continuing to grow, there is high variability in the amount of lateralization provided by the majority of rTSA implants currently available.⁵⁶

With the growing popularity of rTSA and rapid advancements, attention has returned to the role of baseplate positioning on outcomes and complications. The purpose of this review is thus to revisit the current state of literature as it relates to glenoid baseplate positioning from both the clinical and biomechanical perspective. We will discuss the importance and implications of coronal placement and inclination, as well as review relevant biomechanical and clinical studies investigating outcomes of variations in baseplate positioning and inclination and discuss how these parameters may be changing with modern implant designs.

Coronal placement

In the original Grammont design and technique, the glenoid baseplate was placed centrally in the glenoid to maximize bony contact.²⁹ Unfortunately, this led to high rates of scapular notching ranging from 44-96%.^7,49 As a way to reduce notching, many have advocated for an inferior placement of baseplate on the glenoid surface. Biomechanical studies have served to demonstrate an increased impingement free range of motion (ROM) with an inferiorly placed baseplate. In a cadaveric model with 4 different baseplate positions (centered, flush with the inferior rim, overhanging the inferior rim, and 15° inferior tilt), Nyffeler et al demonstrated an increased abduction and adduction angle for shoulders with a glenosphere flush with the inferior glenoid rim and an even further increase with overhang.⁴³ Computer modeling studies evaluating impingement free ROM performed by Gutierrez et al as well as Roche et al demonstrated similar findings, suggesting that placement of the baseplate flush with the inferior rim or with overhang resulted in improved impingement free adduction.^11,26

Clinical studies have shown similar findings. In particular, Dean et al retrospectively reviewed 284 rTSA to evaluate functional outcomes as well as rates of notching. This study found increasing overhang of the prosthesis over the inferior glenoid rim was associated with both improved active forward elevation as well as decreased notching rates.¹⁶ In a similar way, Collotte et al retrospectively compared patients with glenospheres flush to the inferior glenoid and those with at least 3.5 mm of overhang. They also found improved outcomes and significantly lower rates of notching in the overhang group.¹³ Neither study identified optimal overhang, though 3.5 mm seems to be appropriate. In addition, Choi et al compared the clinical and radiological results of 20 reverse shoulder arthroplasties using an eccentric glenosphere to those using a concentric glenosphere.¹⁰ While clinical outcomes were not significantly different in the short-term follow-up, the eccentric glenosphere was more effective in reducing the rate of notching than the concentric glenosphere.

It is important to note that too inferior placement of the baseplate may result in insufficient bone stock for adequate fixation. One way to overcome this concern is with the use of an eccentric glenosphere, which allows for central placement of the baseplate with inferior glenosphere coverage. A computed tomography (CT) scan based study by Werner et al demonstrated decreased scapular notching rates with the eccentric glenosphere.⁵⁴ Several clinical Studies have confirmed the ability of the eccentric glenoid to decrease notching rates compared to the concentric glenosphere as well as improve clinical outcomes.^4,13,38 Keep in mind, notching may still occur with the eccentric glenosphere without appropriate inferior coverage of the glenoid.⁴¹ This was again shown in a randomized controlled trial by Poon et al investigating the rates of notching and clinical outcomes between eccentric and concentric glenosphere placed at the inferior border of the glenoid. This study demonstrated no significant difference between the 2, and noted the most predictive factor to decrease notching was overhang of >3.5 mm.⁴⁸

Although the benefits of inferior placement are clear regarding notching, the concern of increased shear at the bone implant interface as a result of inferior placement remains unanswered. Poon et al first demonstrated increased micromotion with the use of an eccentric glenosphere compared to concentric glenosphere in a biomechanical analysis, though the 26.83 μm is well below the threshold to impair bony ingrowth.⁴⁷ In a similar fashion, strain amounts seen with inferior placement and eccentric glenosphere did not reach levels to disrupt bony ingrowth in a hydraulic testing apparatus.⁴⁵ In a finite element analysis, Zhang et al demonstrated increased bone resorption with inferior placement but no difference in shear strain.⁵⁷ No clinical studies with relation to baseplate loosening and inferior placement were available to include in this review.

Now, as more and more surgeons are lateralizing the glenosphere in rTSA, the idea of inferior glenosphere placement is being questioned as this may adversely affect deltoid muscle force. Some surgeons are placing the glenosphere more superiorly and closer to the midpoint to maintain esthetic shoulder contour as well as to minimize distalization of the arm. However, there is no literature to date that specifically investigates the effects of baseplate placement in the setting of a lateralized glenospheres. In addition, confounding this is the transition to a more anatomic neck shaft angle on the humeral side leading to less distalization and rates of scapular notching independent of the glenoid changes.⁴² The downside of inferior overhang is that it lengthens the arm and distalizes the humerus.^1,34 This has cosmetic issues and potentially may lead to excess tension on the conjoint tendon and deltoid. With a lateralized glenosphere, impingement-free motion can be achieved by having the center of rotation move away from the glenoid face.¹⁴ As such, a lateralized glenosphere can hypothetically be placed more superiorly on the glenoid face. Whether the glenosphere can be placed in the center of the glenoid, or just flush with the glenoid surface, is currently unknown. However, putting a lateralized glenosphere with inferior overhang likely is not necessary and may cause too much tension on the construct.

Version

A decisive consensus has not yet been reached on the optimal degree of glenoid version in rTSA. The maximal amount of anteversion described for reverse is with the alternate center line technique by Frankle and colleagues, which is defined as a line aiming at the dense bone where the scapular spine meets the body of the scapula and is not necessarily perpendicular to the remaining glenoid face.⁵³ Frankle et al conducted a clinical study to classify glenoid morphology and evaluate the effect on glenoid component fixation in the context of identifying the optimal center line.²² They found that in the normal glenoids, the spine centerlines were anteverted 9.2° ± 4.9° from the standard centerlines. In the abnormal glenoids, the spine centerlines were anteverted in each group as follows: 20.2° ± 10.4° (P < .0001) in the posterior erosion group, 8.6° ± 6.3° in the superior erosion group, and 6.4° ± 5.6° (P = .05) in the global erosion group.²² More clinically, Colley et al performed a retrospective case-controlled study to compare outcomes of rTSA using the anatomic or alternative center line.¹² They concluded the alternative center line can be used for baseplate fixation in the setting of glenoid bone loss and leads to similar patient outcomes and functional tasks of internal rotation, as well as a low rate of complications, compared with the anatomic center line following rTSA.

There is ongoing evidence to suggest that restoring glenoid version to around 0° leads to most optimal outcomes. Berton et al sought to review current literature regarding the implications of both humeral and glenoid version as well as provide recommendations on the most optimal degree of humeral and glenoid component versions for rTSA.³ One of the included studies tested 20 patients with cuff tear arthropathy and glenoid retroversion greater than 15°. The mean preoperative retroversion of the glenoid was 24°, while the postoperative was 2°.³³ With regards to clinical outcomes, postoperative Constant scores and Subjective Shoulder Value scores showed a statistically significant increase in 10 points at both 6-month and 12-month follow-up, respectively.³² In another study, investigators looked at how variations in glenoid version influenced arm abduction angle, possibly resulting in impingement.⁴⁶ The study was performed on 35 computer models taking into consideration 5° anteversion; neutral; and 5°, 10°, and 20° retroversion of the glenoid component.⁴⁶ The authors found that retroversion reduces the likelihood of subluxation, concurrently increasing external rotation and overall shoulder motion. In a third study, Favre et al assessed the influence of both humeral and glenoid component versions on rTSA stability to identify parameters that could decrease dislocation rates.¹⁹ They assessed glenoid version in 10° increments from 20° retroversion to 20° anteversion in the 90° abducted and resting positions, finding that only 20° retroversion led to a statistically significant drop in stability (when arm positioning at 20° abduction to stimulate hanging-arm or resting position). Authors concluded that the glenoid component version is less important for intrinsic stability compared to humeral component version, but special care should still be taken to avoid glenoid retroversion greater than 10°.¹⁹ Ellmallah et al evaluated the correlation between preoperative and postoperative retroversion in 271 rTSA patients and determined the effect of glenoid retroversion on functional outcomes, ROM, and postoperative complications.¹⁸ They found no significant difference in American Shoulder and Elbow Surgeons, visual analog scale or Single Assessment Numeric Evaluation scores, or ROM or complications between patients who had a baseplate retroversion less than 15° versus those who had retroversion greater than 15°. More recently, Galasso et al retrospectively reviewed 274 patients while grouping them by postoperative retroversion to <10°, 10-19°, and >20°. They found patients with >10° of retroversion had significantly improved internal rotation at 90° without sacrificing external rotation compared to those with <10° of retroversion.²³ While these studies provide evidence of good outcomes based on evaluating different versions, there is no study that has shown guidelines for version of the baseplate and its' associated long-term outcomes.

Inclination

The inclination of the glenosphere is the angle of the glenoid baseplate as viewed in the coronal plane in the relation to the true scapular plane.³⁰ This has classically been measured with the beta angle, which is defined as the angle created between the floor of the supraspinatus fossa and the glenoid fossa.⁴⁰ Although the beta angle can be calculated with use of conventional radiography, the use of two-dimensional CT scan improves the reliability of the measurement.¹⁵ It is important to note that the beta angle takes into account the entire glenoid fossa, while the reverse baseplate only occupies the inferior aspect of the glenoid to prevent notching. For this reason, some authors have suggested the use of a new angle to measure glenoid inclination in the rTSA named the rTSA angle. This angle specifically factors in the inferior inclination of the glenoid in relation to the supraspinatus fossa and has been demonstrated to be reliable and reproducible.^6,55 Previous studies have made it apparent that the inclination of the glenoid can significantly impact the outcomes after rTSA.

Biomechanical studies on inclination

Although multiple reports suggest that an inferior inclination of the glenosphere improves overall long-term stability of rTSAs, controversy still exists regarding the influence of inferior tilt on glenoid components. One recent study suggested that a 15° inferior tilt of the glenoid baseplate improved the stability and reduced likelihood of mechanical failures.²⁷ In contrast, Nyffeler et al showed that placing the glenoid component at an inferior tilt alone was not as effective in preventing inferior impingement compared to placing the component without a tilt but in a more inferiorly translated position of the glenosphere relative to the glenoid.⁴³ In addition, he showed that the range of shoulder motion was limited to a greater extent in the study specimens that had a glenoid component inferiorly tilted. Chae et al chose to evaluate fresh-frozen cadavers of female individuals greater than 60 years of age due this patient population most commonly experiencing cuff tear arthropathy.⁹ They aimed to evaluate the effects of inferior tilt on micromotion and loss of fixation of the glenoid component, with findings demonstrating that micromotion was greater with a 10° inferior tilt compared to neutral tilt at various cyclic loads. In another study, Chae et al found that eccentric reaming (with efforts to produce the inferior tilt fixation of the glenoid component) increased glenoid cancellous bone exposure and decreased bone–screws contact area.⁸ His work proposed that inferior tilt fixation of the glenoid component may adversely affect primary stability and longevity after rTSA.

As such, current biomechanical literature offers conflicting recommendations for glenoid component inclination as these studies are not without limitations. Originally, biomechanical studies showed that glenoid baseplate had less shear force if they were placed in inferior tilt in relation to the glenoid face.²⁸ Other biomechanical studies advocating inferior tilting of the glenoid component to enhance glenoid fixation include the use of solid rigid polyurethane foam or sawbone blocks to simulate the mechanical properties of the glenoid bone, and the use of a 15° tilt relative to the flat surface of the block.26, 27, 28 Furthermore, some studies proposing inferiorly tilted placement of the glenoid component used computer simulation, which does not consider the anatomic variations or material characteristics of a true glenoid bone.²⁶ Conclusions from these studies should be interpreted judiciously given the fact it is saw bone and computerized data. Although most of these studies solely studied impingement free ROM and rates of notching, a more recent study by Knighton et al utilizing cadaver models demonstrated increased shear force at the baseplate/bone interface with superior tilt by 125% as compared to inferior tilt. In addition, inferior tilt produced 25% higher compressive forces.³² These studies all show that inferior tilt is primarily a biomechanical theory to make baseplate failure less common. With modern implants, it is unknown if this theory still holds true or if some level of superior tilt is acceptable.

Clinical studies on inclination

A majority of clinically based recommendations for placing the glenoid component with some inferior tilt were originally based on retrospective observations, however these studies did not explicitly investigate the effects of glenoid placement and tilt.^21,36,50 In addition, inferior tilt was not significantly different between 2 groups. Other studies have used clinical data and imaging in attempts to simulate various glenoid configurations and risk of scapular notching. Patel et al recently examined the CT scans of 20 patients without glenoid bone loss that underwent rTSA.⁴⁴ They simulated impingement-free ROM with 16 different implant configurations, with every configuration demonstrating that inferior tilt led to more impingement on the scapular neck. Glenoid inferior tilt of −10° was also found to be associated with 27% decrease in impingement-free external rotation and 32% decrease in impingement-free adduction. These findings were attributed to the increased medialization required to seat inferiorly tilted implants, consequently shortening the scapular neck and reducing the distance between the humerus and scapula.

In efforts to establish recommendations based on prospective clinical data, Edwards et al conducted a randomized controlled trial evaluating if 10° of inferior tilt of the glenoid component can decrease radiographic scapular notching as well as examining the influence of inferior tilt of the glenoid component on outcomes and mobility.¹⁷ They found no significant difference in rates of scapular notching or clinical outcomes between those with the glenoid component inferiorly tilted 10° versus those with no inferior tilt. These results were consistent with findings from previous biomechanical studies mentioned,²⁶ however it is difficult to compare results with predictions of biomechanical models which comprised multiple various combinations of implant or surgical factors. However, this is the first prospective randomized controlled trial to date examining the influence of inferior tilt of the glenoid component on radiographic and clinical outcomes after reverse shoulder arthroplasty.

More recent retrospective studies have added some cloudiness to the inclination debate. Tashjian et al retrospectively reviewed 97 patients undergoing rTSA, with 13 reporting instability and 4 patients suffering dislocation events. On multivariate analysis, only superior glenosphere inclination, as measured by beta angle, was found to be significantly associated with instability. Instability was documented based on a patient reported questionnaire.⁵² In a separate study but similar design, Tashjian et al also demonstrated that superior inclination was associated with increased postoperative pain, but no difference in functional outcomes.⁵¹ Both these studies utilized Grammont style prosthesis. On the other hand, Bechtold et al used a lateralized glenoid design and matched 34 cases of instability to 102 controls then compared beta angles between the 2 groups. This study found no association between no statistically significant link between final beta angle or change in beta angle and prosthetic instability.²

In a similar fashion, Mahendaraj et al retrospectively evaluated the rTSA angle in a consecutive series of 154 patients undergoing rTSA with a lateralized system. This study found no difference in ROM, patient reported outcomes or complication rates between inferior inclined, neutral, and superior inclined glenospheres up to 6° of superior inclination. In addition, they demonstrated that postoperative inclination was significantly associated with preoperative inclination.³⁹ In particular, this study challenges the results of prior studies that have utilized inconsistent measures of inclination. In addition, this study utilized a lateralized design. It is possible that increased lateralization decreases the importance on glenoid inclination on outcomes. In the current landscape of rapidly evolving rTSA prosthetic design, it may be important to revisit the role of glenoid inclination on outcomes particularly for the lateralized glenoid.

Glenoid baseplate augments allow for avoidance of superior tilt without having to lose bone and can correct glenoid deformity while potentially avoiding certain complications encountered with structural bone graft.³¹ However, the long-term clinical outcomes of this are not known. Kirsch et al performed a retrospective review of all patients with an augmented baseplate during primary rTSA with a minimum of 1 year clinical and radiographic follow-up, concluding that primary rTSA with an augmented baseplate results in excellent short-term clinical outcomes and significant deformity correction in patients with advanced glenoid deformity. There were no complications related to the augmented baseplate or glenoid component. The rate of acromial stress fractures appears higher than typically reported and warrants further investigation.³¹ Gulotta et al completed a comparative study of patients who underwent rTSA with 8° posterior, 10° superior, or a combination of 8° posterior/10° superior augmented glenoid baseplates with patients who received a standard baseplate.²⁵ They found that at an average follow-up of nearly 4 years, the augmented baseplates performed as well, or better, than standard baseplates, with a similar complication rate, revision rate, aseptic glenoid loosening rate, and scapular notching rate.

Conclusion

Glenoid baseplate positioning continues to evolve as we have yet to establish a clear consensus on the gold standard of placement. Inferior placement of the glenoid may not be necessary based on the growing popularity of lateralized glenoid design. Inferior glenoid tilt is biomechanically stronger; however, the use of a lateralized glenoid may negate the negative effects previously seen with superior glenoid tilt. There remains little knowledge regarding the optimal glenoid version. Given the rapid design developments of the rTSA, the importance of glenoid baseplate positioning on functional outcomes may need to be revisited.

Disclaimers

Funding: No funding was disclosed by the authors.

Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.