28.1 Introduction

Polyethylene glenoid component is the gold standard in anatomic total shoulder arthroplasty (ATSA). Although the survivorship of the anatomic shoulder prosthesis is more than 95% at 10-year follow-up, ¹ glenoid component loosening is one of the main complications in ATSAs.

This complication represents 25% of all complications related to ATSA. ² In most cases, glenoid component loosening is not isolated but combined with a rotator cuff tear, glenohumeral instability, or component malposition. ³ , ⁴ , ⁵ , ⁶

Overall, revision of an ATSA to a reverse total shoulder arthroplasty (RTSA) has required the removal of both the humeral and glenoid components. Removal of the humeral stem sometimes requires removing the cement, osteotomy of the diaphysis with the risk of fracture, and extensive damage of soft tissue. ⁷ , ⁸ , ⁹ Removal of the glenoid component with cement often leaves a huge glenoid bone loss with substantial problems in fixation of a new glenoid baseplate. Allograft and special implant design with a long central post are mandatory to obtain a stable fixation of the new baseplate in these difficult cases. ¹⁰ , ¹¹ , ¹² , ¹³

With a complete convertible platform system on humeral and glenoid side, during the revision of a failed ATSA, one can leave the humeral stem and the glenoid baseplate in place. On the humeral side, the humeral head is replaced by the appropriate humeral tray component, and on the glenoid side, the glenosphere is impacted onto the glenoid baseplate.

Our hypothesis was that a complete convertible system on both sides would facilitate the revision of ATSA to RTSA avoiding the difficulties to reimplant a glenoid component particularly on osteoporotic glenoid bone and elderly patient. The purpose of this chapter is to outline the feasibility of a convertible glenoid system and to report the early clinical results of a retrospective review of 10 cases.

This chapter will outline the use of convertible implants for glenoid reconstruction in revision of failed ATSA to reverse shoulder arthroplasty utilizing a convertible platform system (► Fig. 28.1, ► Fig. 28.2).

28.2 Indications

Patients with failed ATSA that require revision often have developed significant glenoid loosening either primarily or secondarily after instability, subscapularis failure, and rotator cuff insufficiency. Often, with glenoid component excision, there is resulting significant glenoid bone stock deficiency, making glenoid reconstruction in revision to RSA quite challenging. The premise of this chapter is that a convertible glenoid platform system which could be revised directly to RSA would make the reconstruction easier and safer for the patient.

Between 2003 and 2011, a total of 104 ATSAs were performed with an uncemented glenoid component as a primary procedure (Arrow; FH Orthopedics, Mulhouse, France; ► Fig. 28.1). On those 104 cases, 16 were lost to follow-up due to death or failure to return. Finally, 82 patients underwent 88 ATSAs with a baseplate glenoid implant. Of these, 14 patients had to be revised (9.8%). The initial ATSA was indicated for 10 primary osteoarthritis with a normal cuff, 2 primary osteoarthritis with a cuff tear, 1 posttraumatic osteoarthritis, and 1 chronic dislocation. There were no previous surgeries in any of these patients, with the exception of a posttrauma patient who underwent open reduction internal fixation. Among these 14 patients, 2 dissociations of the PE tray and 2 low implantation of the baseplate were revised with a new ATSA and 10 arthroplasties were converted from an ATSA to a RTSA using the modular convertible Arrow system (FH Orthopedics).

In this retrospective study of 10 patients with a mean age of 64 years (range: 50–75)—eight women and two men—the causes of the conversion to RTSA were one glenoid loosening, three rotator cuff tears (► Fig. 28.3), four dislocations (three biconcave glenoid type B2), and one suprascapular nerve paralysis.

The right shoulder was involved in eight cases and the left in two with the dominant side in seven cases. The mean time interval from the primary ATSA to revision was 15 months (range: 1–61 months).

28.3 Preoperative Planning

We systematically performed standard X-rays (AP, Axillary, Outlet view) and computed tomographic (CT) scans in order to evaluate the version, the positioning, any signs of loosening of the implants, and the status of the cuff (continuity, trophicity, and degree of fatty infiltration of the muscles).

28.4 Surgical Technique

All these revision surgeries utilizing convertible implants were performed through the previous deltopectoral approach with the patient positioned in beach chair position under general anesthesia with an interscalene block. First, all the subdeltoid adhesions were released sharply and with electrocautery. The conjoint tendon was then identified and the location of the musculocutaneous and axillary nerves were palpated and identified with the index finger before the retractors were placed. The subscapularis was released laterally at the medial border of the bicipital groove to gain length and avoid a repair under tension later.

Next, the humeral head component of the anatomic shoulder prosthesis was removed from the humeral stem. In this series all the humeral stems were found to be well-fixed with no cases of loosening or infection. The polyethylene onlay component on the metal backed base plate on the glenoid side was then removed. A circumferential capsular release was carried out.

If an infection was suspected, at least five culture biopsies were taken in various areas to diagnose the infection and procure the pathologic organism. The quality of the fixation of the glenoid base plate was then evaluated for stability. With this convertible system there was no need to exchange the glenoid base plate even in cases where the position of the baseplate was considered too retroverted (Glenoid B2) or too high.

At this time a convertible glenosphere component of appropriate diameter and offset was then positioned and fixed into the retained convertible baseplate. All the humeral stems were well fixed and so at this point an onlay polyethylene humeral socket was impacted on the humeral stem. In most cases in this series thinner humeral polyethylene sockets and 36 mm glenospheres were utilized because of the possibility of inability to reduce the components or over-tightening the joint with bigger implants.

In one case the humeral stem was a first generation type humeral implant, which was not convertible in the present system.

In two other cases, it was impossible to reduce the components in the revision to the reverse total shoulder arthroplasty even after an extensive soft tissue release and resection of the supraspinatus tendon. The most common reason for this situation was that the excessive height of the well fixed humeral stem compromised the space necessary for the humeral cup insertion.

With this convertible shoulder arthroplasty system, removal of the non-cemented humeral stem was always possible without osteotomy of the humerus shaft or distal window. Once the component had been removed, some of the proximal humeral metaphysis was resected with a reciprocating saw and a new humeral stem was then impacted in lower position making reduction of the revision components possible. In this retrospective study, no pre- or postoperative complications were observed. The mean duration of the procedure was 60 minutes (range 30 to 75). Patients were hospitalized for a mean of 4 days (range 3 to 6; ► Fig. 28.4, ► Fig. 28.5, ► Fig. 28.6, ► Fig. 28.7, ► Fig. 28.8, ► Fig. 28.9, ► Fig. 28.10).

After anatomic total shoulder arthroplasty with a subscapularis tear and anterior/inferior instability, neither repair nor tendon transfer was attempted, instead reverse shoulder Arthroplasty was performed. We prefer to lateralize the prosthesis to increase deltoid tension. In this series no postoperative instability was observed in any patient (► Fig. 28.11, ► Fig. 28.12).