Reverse total shoulder arthroplasty (RTSA) inverts the native glenohumeral joint articulation to restore stability to an otherwise unstable glenohumeral joint. This is achieved by making the glenoid articulation convex and the humeral articulation concave to create a fixed fulcrum that geometrically prevents superior migration and acromial contact as the deltoid contracts to elevate the arm. Inverting the concavities inferiomedially translates both the center of rotation (CoR) and humeral position relative to that native shoulder (FIGURE 4.1). While RTSA prostheses vary in the degree of CoR and humeral translation, contemporary designs are associated with a 5 to 10 mm inferior¹,²,³,⁴ and 20 to 30 mm medial²,³,⁵,⁶,⁷,⁸ translation of the CoR and a 25 to 40 mm inferior²,³ and 5 to 20 mm medial²,³ translation of the humerus relative to the native shoulder. Medially translating the CoR has the added benefit of increasing the length of the deltoid abductor moment arm from 10 to 30 mm⁹,¹⁰,¹¹,¹²,¹³ for the native shoulder with the arm at the side to 22 to 40 mm¹³,¹⁴,¹⁵,¹⁶ for the RTSA construct (FIGURE 4.2). This increase in the length of the abductor moment arm improves the efficiency of the deltoid by requiring less muscle force to generate the same amount of torque.

Unlike the native shoulder, which has a nonconforming articulation that permits humeral head rotation, translation, and rolling on the glenoid, the RTSA articulation is conforming, so its motion is limited to humeral rotation without translation (FIGURE 4.3). In the native shoulder, the rotator cuff acts as a dynamic fulcrum that facilitates arm elevation by pivoting the humeral head against the glenoid as the deltoid contracts.¹⁷ By comparison, in the RTSA shoulder, the inverted concavity converts the superiorly directed deltoid muscle vector into arm elevation and rotation. The magnitude of motion achieved with RTSA in each anatomic plane is dependent upon several factors including humeral liner constraint, humeral liner-scapular impingement, humeral neck angle, humeral retroversion, glenosphere inferior overhang, patient anatomy/morphology, and available musculature.¹,²,³,¹⁵,¹⁶,¹⁸,¹⁹,²⁰,²¹,²²,²³,²⁴,²⁵,²⁶,²⁷,²⁸,²⁹,³⁰,³¹,³²,³³,³⁴,³⁵,³⁶,³⁷,³⁸

Regarding the musculature required for active range of motion (ROM) with RTSA, a functioning deltoid is necessary for arm abduction, forward flexion, and also for joint stability. A nonfunctioning deltoid is a contraindication of RTSA. Internal rotation is achieved by the action of the subscapularis, teres major, pectoralis major, latissimus dorsi, and/or anterior deltoid. A functioning teres minor, at a minimum, is necessary to achieve active external rotation, and while external rotation improvement after RTSA may be unpredictable, most patients will achieve more active external rotation if they also have a functioning infraspinatus. If neither the teres minor nor infraspinatus is functional, the posterior deltoid may contribute to external rotation, but a muscle transfer of one of the large internal rotator muscles, most commonly the latissimus dorsi, may also be necessary.

The RTSA prosthesis was first designed in the 1970s; several prostheses were developed, including the Fenlin, Reeves-Leeds, Kessel, and Neer-Averill shoulders.³⁸,³⁹,⁴⁰,⁴¹,⁴²,⁴³,⁴⁴,⁴⁵ Each of these prostheses had a constrained and conforming articulation whose CoR was lateral to the glenoid fossa. These design features were associated with excessive torque on the glenoid bone-implant interface that compromised fixation and resulted in aseptic glenoid loosening and/or mechanical failure. Due to the high implant failure rates, the early RTSA designs were abandoned in the US market.⁴¹,⁴²

Interest in RTSA reemerged when Dr. Paul Grammont introduced his Delta III reverse prosthesis in Europe in 1991 and in the United States in 2003. The Delta III RTSA prosthesis utilized a hemispherical glenosphere whose thickness was equal to its spherical radius to medially position the CoR directly on the glenoid fossa.⁶,⁴¹,⁴⁶,⁴⁷,⁴⁸ By medializing the CoR, the Grammont RTSA prosthesis minimized torque on the glenoid bone-implant interface to reduce the incidence of aseptic glenoid loosening, while also increasing the abductor moment arm length to improve deltoid efficiency and reduce the force required to elevate the arm. This medialized CoR concept is Dr. Grammont’s great innovation that all contemporary RTSA designs generally share. These contemporary designs have been associated with predictable improvements in
function and stability for a variety of difficult-to-treat degenerative conditions of the shoulder including cuff tear arthropathy, osteoarthritis with rotator cuff tears, proximal humeral fractures, arthroplasty with glenoid bone loss, arthroplasty with proximal humeral bone loss, and also revision arthroplasty.⁴⁸,⁴⁹,⁵⁰,⁵¹,⁵²,⁵³,⁵⁴,⁵⁵,⁵⁶,⁵⁷,⁵⁸,⁵⁹,⁶⁰,⁶¹ Market response to these positive clinical experiences has dynamically altered the usage pattern of shoulder arthroplasty in the United States, as characterized by a sharp increase in RTSA since its introduction, a significant decline in hemiarthroplasty, and a relative reduced growth of anatomic total shoulder arthroplasty (ATSA) (FIGURE 4.4).

Medializing the CoR with the Grammont RTSA prosthesis inferiomedially translates the humerus and alters the native orientation of the humeral muscle insertions relative to the CoR. Doing so changes each muscle’s moment arms and native lengths and modifies how each muscle influences motion relative to its native physiologic function. This inferiomedial humeral positioning has multiple negative biomechanical implications, including elongation of the deltoid by as much as 20%,¹,²,³,⁵,¹⁴,¹⁵,¹⁶,³²,³³,⁶²,⁶³ reduced deltoid wrapping around the greater tuberosity,²,³,⁷,⁹,³¹,³⁴,⁶⁴,⁶⁵ reduced ROM and impingement of the humeral liner with the scapular neck (leading to scapular notching),¹⁸,²⁰,²¹,²²,²³,²⁴,²⁹,³²,³³,³⁵,³⁶,³⁷,⁵⁸,⁶⁶,⁶⁷,⁶⁸,⁶⁹,⁷⁰ and shortened rotator cuff muscle lengths, which impair their ability to generate internal/external rotation.²,³,³⁴,⁷¹,⁷²,⁷³ Shortening of the rotator cuff muscles⁷³ may be responsible for the modified scapulohumeral rhythm with RTSA, where additional scapular rotation occurs relative to the native shoulder, potentially as a compensatory mechanism for the lax scapulohumeral muscles during elevation⁷⁴ (FIGURE 4.5).

RTSA biomechanics can be altered by prosthesis design parameters. Since the Delta III RTSA prosthesis was introduced in the United States in 2003, numerous RTSA prostheses have been developed, many of which
incorporate design improvements that address the aforementioned biomechanical limitations of the Grammont prosthesis. Specifically, RTSA prosthesis design changes sought to (1) preserve the amount of bone removed during implantation⁷⁵; (2) enhance glenoid fixation and provide more options for glenoid bone loss in primary and revision procedures⁷⁵,⁷⁶,⁷⁷,⁷⁸,⁷⁹,⁸⁰,⁸¹,⁸²,⁸³; (3) improve ROM and reduce scapular notching¹⁸,²³,²⁴,²⁸,²⁹,³²,³³,³⁵,³⁶,⁶⁶,⁷⁰; and (4) lateralize the humerus to further increase abductor moment arms,¹,⁵,¹³,¹⁴,¹⁵,¹⁶,³¹,³⁴,⁷¹ restore rotator cuff muscle length,²,³,³⁴,⁷²,⁷³ and restore deltoid wrapping.²,³,³⁴ These design modifications have largely resulted in positive clinical improvements by reducing the occurrence of certain complications such as instability,³⁴,⁸⁴,⁸⁵,⁸⁶ acromial and scapular fractures,⁸⁷,⁸⁸,⁸⁹,⁹⁰,⁹¹ and the incidence of scapular notching.¹⁸,²³,²⁴,²⁸,²⁹,³²,³³,³⁵,³⁶,⁶⁶,⁷⁰ However, not all of these design changes have resulted in biomechanical improvements.

To mitigate scapular notching and improve ROM, several RTSA prostheses have attempted to lateralize the humerus by lateralizing the CoR through either use of thicker glenospheres⁴⁹,⁵¹ or by placing bone graft behind the glenoid baseplate.⁵³,⁵⁴,⁵⁵,⁹² Lateralizing the CoR has been demonstrated to reduce humeral liner impingement with the scapular neck.¹,³²,³³,⁴⁹,⁵¹ However, lateralizing the CoR also increases the torque on the glenoid bone-implant interface (which can compromise fixation)⁷⁸,⁸²,⁹³ and decreases the length of the deltoid abductor moment arms (which reduces deltoid efficiency).¹,¹³,¹⁴,³⁴,⁶⁵,⁶⁶,⁷¹ As the deltoid abductor moment arms are decreased, the deltoid becomes less effective as an abductor and requires a greater force to elevate the arm.¹,¹³,¹⁴,³⁴,⁶⁵,⁶⁶,⁷¹ Additionally, these elevated muscle loads increase scapular bone stresses, thereby, increasing the risk of acromial and scapular insufficiency fractures, which occur at a higher rate in RTSA prostheses that lateralize the CoR relative to other design styles.⁸⁸,⁹⁴,⁹⁵,⁹⁶