Staged Revision Reverse Total Shoulder Arthroplasty to Functional Hemiarthroplasty With Retained Humeral Implant: A Salvage Technique

Abstract

With the increased incidence and indications for reverse total shoulder arthroplasty, the increased need for revision shoulder arthroplasty also continues to rise. Therefore, revision strategies and techniques are essential to optimize patient outcomes and preserve function, particularly in the challenges of glenoid failure, poor glenoid bone stock, and/or poor native bone quality. Because of the risk of intraoperative fracture, increased blood loss, prolonged operative time, excessive implant costs, and the high risk of complications of a humeral stem extraction and revision, extraction of a well‐fixed humeral stem may be unnecessary and undesirable. When feasible, stem retention is preferred, which can lead to reduced surgical morbidity, cost‐saving practice habits, and better patient outcomes, especially in systems lacking platform convertibility. This Technical Note expands the reconstructive salvage options in systems lacking platform convertibility through conversion of a reverse total shoulder arthroplasty to a functional hemiarthroplasty by cementing a glenosphere onto a retained, nonconvertible, stemmed humeral tray. This technique may serve as a definitive solution or, more likely, a temporary construct in select salvage and revision cases, offering joint stability and pain control while preserving articular congruency and avoiding the morbidity associated with full‐component revision.

Reverse total shoulder arthroplasty (rTSA) makes up nearly 70% of all total shoulder arthroplasty performed in the United States, and use continues to rise as indications extend beyond rotator cuff deficiencies to include primary glenohumeral osteoarthritis, severe glenoid deformity, and bone loss, as well as revision settings. ¹ , ² With over 800,000 Americans currently living with an rTSA, and a projected growth rate exceeding 700% by 2040, the need for revision options is inevitable and expected to increase substantially. ² , ³ , ⁴ , ⁵ As a result, understanding and improving revision strategies are critical for the treating surgeon, particularly in the setting of persistent postoperative pain, impaired function, osteolysis, instability, infection, or implant failure after rTSA. ⁴ , ⁶ , ⁷

Instability is a major cause of rTSA failure, responsible for 30%‐50% of all revision cases, which can be because of underlying glenoid component failure, humeral loosening, glenoid or humeral bone loss, superior inclination of the glenosphere, and/or fracture. These complications can lead to significant pain, functional limitations, and low patient satisfaction. ⁴ , ⁶ , ⁷ , ⁸ When revision is required, a revision rTSA is likely preferred as it results in superior patient‐reported outcomes and greater maintained range of motion. However, revision rTSA is not always possible, particularly in the setting of insufficient glenoid bone stock, loose implants, and/or active infection, in which conversion to a hemiarthroplasty is likely the preferred salvage procedure. ⁵ , ⁹ , ¹⁰ In salvage situations, a staged revision approach allows time for bone healing and recovery, infection eradication, and/or planning for custom implants. This Technical Note describes the first stage of a salvage option for revision rTSA to a temporary functional hemiarthroplasty using a retained, nonconvertible, stemmed humeral tray (Figure 1).

FIGURE 1.

Preoperative radiographic evaluation including (A) Grashey and (B) axillary views of a right shoulder showing reverse total shoulder arthroplasty with glenoid failure including broken screws, central glenoid bone loss, and superior inclination of the glenosphere, with an overall well‐fixed humeral stem without obvious lucency or malposition.

SURGICAL TECHNIQUE

Patient Positioning and Surgical Approach

The patient is placed in the beach‐chair position with the head elevated to approximately 40° under general anesthesia. The operative extremity is positioned free on a well‐padded Mayo stand. Fluoroscopy is available as needed for intraoperative assessment, positioned behind the head and operative shoulder. A standard deltopectoral approach is used just lateral to the coracoid and through the deltopectoral interval, incorporating the previous incision. In the revision setting, care is taken to perform a careful dissection with meticulous hemostasis because of the distorted anatomic landmarks, which are often encased in scar. If present, the cephalic vein is mobilized laterally.

Subscapularis Management and Axillary Nerve Neurolysis

If the subscapularis is present, either a peel or a tenotomy is performed to access the glenohumeral joint. In this case, we use a peel based on surgeon preference. In the setting of scar tissue and adhesions, the capsule is often incorporated and unable to be separated from the overlying subscapularis. Five intraoperative samples for frozen section are obtained once the glenohumeral joint is entered for high‐powered field evaluation of periprosthetic joint infection. It is critical to identify and protect the axillary nerve during the deep approach in the revision setting. Once identified, the axillary nerve can be carefully released from the surrounding fascia to aid with appropriate visualization without excessive retraction on the nerve. The nerve is carefully mobilized medially around the anterior glenoid to decrease the risk of scar entrapment and deltoid dysfunction postoperatively.

Removal of Retained Glenosphere and Humeral Components

A sharp Hohmann retractor is used to dislocate the joint within the interval between the polyethylene liner and glenosphere. The humeral tray is fully exposed with the extremity adducted, and the polyethylene insert is removed from the humeral tray. The humeral stem, as well as the proximal humeral bone stock, is carefully interrogated and can be retained in place if well‐fixed without any instability or motion. The humeral tray can either be exchanged at this point in the case or can be retained and exchanged at the second‐stage procedure.

The glenosphere is next exposed using a Fukuda retractor laterally, a glenoid neck retractor medially, and a pointed Hohmann retractor superiorly. After removal of the central screw and glenosphere, the native glenoid‐baseplate interface is interrogated for stability. If unstable—including gross micromotion, broken hardware, and extensive central bone loss—the glenoid baseplate is carefully removed en bloc in a bone‐preserving fashion with the combination of rongeurs, curettes, and osteotomes as needed. Any broken hardware that is embedded deep within the bone is typically retained to preserve as much bone stock as possible.

Evaluation of Glenoid Vault and Bone Cementing

In cases with large central cavitary bone loss of the glenoid vault, the bone quality is often too poor and too limited to achieve purchase for a new central baseplate screw path in a single‐stage surgery. Therefore, at this point in the case, the decision is made to proceed with a 2‐stage revision with temporary conversion to a functional hemiarthroplasty while retaining the humeral stem. A large, uncontained central cavitary defect of the glenoid vault can be filled with a FastThread biocomposite interference screw (Arthrex, Naples, FL) to prevent cement extravasation through the posterior scapula. The size of the screw should match the diameter of the previous screw or post. This is followed by filling the cavitary defect with Cobalt Bone Cement (Enovis, Wilmington, DE). The native glenoid surface is carefully contoured with bone cement to create a smooth, articulating surface for the functional hemiarthroplasty. Bone cement is used in the glenoid rather than bone grafting because of surgeon preference and for cost‐saving purposes. The bone cement serves as a temporary solution in preparation for a custom glenoid component in the future, which requires advanced imaging without any retained implants to plan.

Conversion to Functional Hemiarthroplasty

When confronted with revision surgery in patients that have undergone rTSA, familiarity with the primary implant used is critical. Based on the initial and retained arthroplasty system (Univers Revers Total Shoulder System, Arthrex, Naples, FL), the humeral stem implant is not easily convertible to a standard hemiarthroplasty. However, modular systems such as this one allow for a solution. Therefore, a temporary, functional hemiarthroplasty is created with a technique of cementing the previously explanted glenosphere into the retained humeral tray (Figure 2). For cost‐saving purposes, the previous glenosphere is retained for this case with planned explantation during the second stage. However, if there is any concern for unrecognized infection, a new glenosphere of the same size can be implanted. Cobalt Bone Cement (Enovis, Wilmington, DE) without antibiotics is mixed on the back table, whereas the undersurface of the glenosphere and humeral tray is rasped to roughen up the metal surface interface. Once ready for implantation, the cement is carefully placed abundantly around the metal undersurface of the glenosphere and within the humeral tray (Figure 3). The glenosphere is then gently impacted into place onto the retained humeral tray and held with continued gentle pressure, whereas additional cement is placed in a 360  fashion around the glenosphere‐humeral tray metal interface. This technique is akin to a grout application—adhering the glenosphere (the new hemiarthroplasty head) to the metal humeral tray. After the cement hardens, the shoulder is reduced, and smooth articulation of the glenosphere with the native glenoid is examined through range of motion (Figure 4).

FIGURE 2.

Intraoperative image of the components required for conversion of a reverse total shoulder arthroplasty to a functional hemiarthroplasty with a retained humeral stem, in the setting of a nonconvertible total shoulder arthroplasty system, including the explanted polyethylene liner, glenosphere, and humeral tray.

FIGURE 3.

Intraoperative images of a right shoulder in the beach‐chair position through a standard deltopectoral interval showing abundant bone cement placement around (A) the undersurface of the glenosphere and (B) within the humeral tray to create the functional hemiarthroplasty. (C) After gentle impaction of the glenosphere onto the humeral tray, additional bone cement is placed around the glenosphere‐humeral tray metal interface.

FIGURE 4.

Immediate postoperative radiographic evaluation including (A) Grashey and (B) axillary views of a right shoulder showing functional hemiarthroplasty after cemented glenosphere onto a retained humeral tray and stem with a concentrically reduced glenohumeral joint and well‐fixed humeral stem.

Closure and Postoperative Rehabilitation

Either the peel or the tenotomy of the subscapularis is repaired along with the capsule. The technique of repair is chosen depending on the soft tissue and humeral stock available. ¹¹ Closure is achieved in a layered fashion with application of vancomycin powder and dilute betadine after deep skin closure. The operative extremity is immobilized in a padded abduction sling for 2 to 3 weeks. Postoperatively, the patient will follow the standard rTSA physical therapy protocol ⁹ to include immediate active and passive range of motion as tolerated.

DISCUSSION

Revision shoulder arthroplasty in the setting of an already revised rTSA is often a salvage procedure complicated by poor bone quality and bone loss, requiring a balance of technical feasibility, biological constraints, cost‐effectiveness, and implant availability. In cases where the humeral stem remains well‐fixed, without concern for infection, extraction may be unnecessary and undesirable, especially in the setting of a prior cemented humeral stem and/or a system that is not an easily convertible platform because of the risk of fracture, increased blood loss, prolonged operative time, and excessive implant costs. When feasible, stem retention is preferred, which has been associated with reduced morbidity and favorable outcomes in revision surgery. ¹²

In this Technical Note, we describe a staged, salvage technique to retain the humeral stem and repurpose the retained humeral tray by cementing a glenosphere directly into it (Video 1). The previously explanted glenosphere is the same one cemented into the retained humeral tray, but it is acceptable to undersize and use additional cement at the metal‐metal interface. This construct—functionally equivalent to a hemiarthroplasty ¹⁰ , ¹³ —allows for soft‐tissue tensioning and preservation of the articular congruency without requiring glenoid reconstruction, allowing the glenoid bone defect to heal. In the event of a second‐stage procedure, the glenosphere is easily removed with an osteotome, and the humeral tray can be swapped for a new component along with the polyethylene liner (Figure 5).

VIDEO 1.

Download video file ^{(88.6MB, mp4)}

This Technical Note describes a revision reverse total shoulder arthroplasty (rTSA) to a functional hemiarthroplasty with a retained humeral implant presented as a salvage technique. This case was performed at The Steadman Clinic in Vail, Colorado. This case involves a 72‐year‐old right‐hand‐dominant man who presents with persistent right shoulder pain with functional limitations in the setting of a prior revision of an anatomic total shoulder arthroplasty to an rTSA for subscapularis failure. On examination, he has limited overhead range of motion. Preoperatively, he has failed conservative therapy and has undergone an extensive infection workup which remains negative. Based on concerns for glenoid failure with a stable cemented humeral stem, the tentative plan was for revision rTSA to a functional hemiarthroplasty for a salvage‐type, 2‐stage revision procedure. Preoperative plain films show evidence of glenoid baseplate failure with screw breakage and superior inclination. There is a large central cavitary glenoid bone loss. However, the previously cemented humeral stem appears intact and well‐seated without any obvious lucency or loosening. The procedure begins with removal of the polyethylene liner after dislocation, with exposure of the humeral tray and careful interrogation of the stability of the humeral stem for retention. Additional consideration is given to the quality of the proximal humerus bone stock. If the humeral stem appears well‐fixed on interrogation, then the decision is made to proceed with retention of the humeral stem and conversion to a functional hemiarthroplasty. At this point in the case, the humeral tray may either be exchanged or retained and exchanged at the planned second‐stage procedure. This can be especially helpful if the humeral stem implant of the retained arthroplasty system is not an easily convertible platform. To create the functional hemiarthroplasty, bone cement is mixed on the back table and abundantly placed around the metal undersurface of the glenosphere and within the humeral tray. The undersurface of the glenosphere and humeral tray can also be rasped to roughen up the metal surface interface to aid with cement interdigitation. Additional cement is placed around the glenosphere‐humeral tray metal interface. The glenosphere is then gently impacted onto the retained humeral tray and held with continued gentle pressure until the cement is completely cured. Additional cement is applied in a 360  fashion to seal the glenosphere to the prior rTSA metal humeral tray. Once dried, the integrity and stability of the glenosphere are thoroughly assessed, showing excellent fixation. The shoulder is then reduced and taken through range of motion to ensure smooth articulation throughout the arc of motion without any mechanical blocks. Immediate postoperative radiographs show a functional hemiarthroplasty after cemented glenosphere onto a retained humeral tray and stem with a concentrically reduced glenohumeral joint. Postoperatively, the patient is immobilized in a padded abduction sling for 2 to 3 weeks for soft‐tissue rest and follows a standard rTSA rehabilitation protocol. During the second‐stage revision procedure, the glenosphere is easily removed from the retained humeral tray using an osteotome and light mallet taps between the metal‐cement interface of the glenosphere and humeral tray. Video content can be viewed at https://doi.org/10.1002/atn2.70295.

FIGURE 5.

Intraoperative images of a left shoulder in the beach‐chair position through a standard deltopectoral interval showing (A) a well‐fixed and positioned glenosphere cemented onto the humeral tray during a second‐stage revision procedure after conversion to a functional hemiarthroplasty. (B) An osteotome and light mallet taps between the metal‐cement interface of the glenosphere and humeral tray are used to easily remove the cemented glenosphere.

A technical consideration in this construct is the direct cementation of a metallic glenosphere onto a metallic tray, resulting in a metal‐on‐metal interface. In this particular rTSA system (Univers Revers Total Shoulder System, Arthrex, Naples, FL), the humeral tray is composed of titanium alloy (Ti‐6Al‐4V), whereas the glenosphere is typically manufactured from cobalt‐chromium. Although cemented fixation of metallic components is well established in total joint arthroplasty, direct bonding between 2 metallic surfaces—without porous coating or potential for bone ingrowth—is unconventional and relies entirely on the integrity of the cement interface for long‐term stability. Although historic concerns about fretting corrosion, micromotion, and long‐term loosening exist in metal‐on‐metal applications, ¹⁴ the mechanical load through the shoulder is much lower than in weight‐bearing joints; furthermore, this technique is not presented as a long‐term solution. Nevertheless, careful surface preparation, appropriate cement technique, and precise component alignment are essential to minimize mechanical complications. Given the lack of precedent in the literature for this specific construct, it should be considered as a temporary, salvage technique to be applied selectively with close clinical and radiographic monitoring over time (Tables 1 and 2).

TABLE 1.

Advantages and Limitations of Staged Conversion of rTSA to a Functional HA

Advantages	Limitations
Avoids morbid stem extraction when the humeral component is well‐fixed—reducing fracture risk, blood loss, and operative time	Appropriate patient selection is critical—outcomes are highly dependent on preserved deltoid function, acceptable soft‐tissue balance, and absence of active infection
Useful in nonconvertible arthroplasty systems—enables a salvage option when platform conversion to an anatomic TSA or HA configuration is not available	Unconventional metal‐on‐metal with cement interface—long‐term fixation depends entirely on cement integrity between metallic surfaces, which risks loosening and micromotion
Provides pain control and stability while preserving articular congruency without immediate glenoid reconstruction—helpful with severe glenoid bone loss or poor bone quality	Functional ceiling—range of motion and strength may be limited compared with a successful revision rTSA, and persistent instability can still occur depending on integrity of soft‐tissue constraints
Facilitates later second‐stage reconstruction—glenosphere can be removed and glenoid components exchanged during definitive revision once optimized	Not a definitive glenoid solution in most cases—typically functions as a staged, temporizing construct while the bone heals and further planning occurs, rather than a durable final reconstruction
Conscious cost‐saving benefits—avoids costly revision implants or commercially available temporary cement‐based components that are temporary during a staged procedure	Limited evidence‐based outcomes—durability, survivorship, and complications are unknown, limiting generalizability

HA, hemiarthroplasty; rTSA, reverse total shoulder arthroplasty; TSA, total shoulder arthroplasty.

TABLE 2.

Pearls and Pitfalls of Staged Conversion of Reverse Total Shoulder Arthroplasty to a Functional Hemiarthroplasty

Pearls	Pitfalls
Plan for revision exposure—extensile deltopectoral approach, meticulous hemostasis, and constant reorientation to the anatomic landmarks in scarred tissue planes	Getting lost in scarred planes—distorted anatomy can lead to bleeding, inadequate visualization, and poor soft‐tissue planes if exposure is rushed
Protect the axillary nerve—identify early and carefully mobilize to avoid traction injury during glenoid work	Axillary nerve traction—excessive retraction or failure to mobilize the nerve in revision cases can risk postoperative deltoid dysfunction and pain
Interrogate the humeral stem thoroughly before committing to retention—retain only if well‐fixed, appropriately positioned, and infection concern is low	Retaining a loose or infected stem—retention without careful stability assessment and high infection suspicion can risk early failure and complicate the next stage
Prevent cement extravasation and shape a congruent glenoid surface—plug any large central vault defects and contour cement to a smooth articulating surface	Poor cement technique/extrusion—extravasation through the glenoid vault or an irregular cement interface can cause pain, mechanical block, instability, and early failure
Maximize metal‐cement‐metal fixation strength—roughen/rasp both the glenosphere undersurface and humeral tray, use abundant cement, and hold alignment until fully cured	Underpreparing surfaces or malalignment during cementation—inadequate surface preparation with poor cement technique can lead to micromotion and debonding

This salvage technique offers the ability to convert to a hemiarthroplasty in the setting of a nonconvertible system after rTSA, while avoiding the morbidity associated with full‐component revision, expanding the reconstructive options in systems lacking platform convertibility. Although this salvage option may serve as a definitive solution in select elderly and low‐demand patients, it also serves as a functional intermediate step in a staged reconstructive strategy, allowing time for bone healing, soft‐tissue recovery, and planning for a future custom glenoid reconstruction to address severe glenoid bone loss or deformity.

DISCLOSURES

The authors (M.T.D., M.T.P.) declare the following financial interests/personal relationships which may be considered as potential competing interests: M.T.D. reports Arthroscopy Journal (editorial board member), American Academy of Orthopaedic Surgeons (board or committee member), and Society of Military Orthopaedic Surgeons (board or committee member), outside of the submitted work. M.T.P. reports American Academy of Orthopaedic Surgeons (board or committee member), American Orthopaedic Society for Sports Medicine (board or committee member), American Shoulder and Elbow Surgeons (board or committee member), Arthrex (IP royalties, paid consultant, and research support), Arthroscopy Association of North America (board or committee member), Arthroscopy Journal (editorial board member), Arthrosurface (IP royalties), Elsevier (publishing royalties, financial, or material support), International Society of Arthroscopy, Knee Surgery, and Orthopaedic Sports Medicine (board or committee member), Joint Restoration Foundation (AlloSource; paid consultant), Knee (editorial or governing board), Orthopedics (editorial or governing board), San Diego Shoulder Institute (Board or committee member), SLACK (editorial or governing board; publishing royalties, financial, or material support), and Society of Military Orthopaedic Surgeons (board or committee member), outside of the submitted work. The other authors (T.A., N.J.D., F.A.U., M.V., M.B.) declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.