March 15, 2018
6 min read

A 70-year-old man with 3 months of worsening left shoulder pain following TSA

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

A 70-year-old right-hand-dominant man presented to the orthopedic clinic with 3 months of worsening left shoulder pain after having undergone total shoulder arthroplasty for primary glenohumeral arthritis 1.5 years earlier at an outside hospital. The patient denied any history of trauma but noted new areas of redness and skin discoloration over his left shoulder. He denied any systemic signs of illness including fevers, chills or night sweats, but did note increasing difficulty with range of motion secondary to pain. The patient attempted 3 months to 6 months of organized physical therapy, but experienced worsening disability and pain with his activities of daily living. The patient was a non-smoker with no significant medical history. On exam, the patient’s left shoulder showed a well-healed deltopectoral incision with small areas of blotchy erythema surrounding the incision. Range of motion (ROM) was 90° forward elevation (FE), 45° external rotation (ER) and internal rotation (IR) to S1. The motor exam demonstrated weakness in FE and IR, but neurologic examination was normal.

Radiographs and CT imaging of the left shoulder showed anterior dislocation of the humeral component with possible increased lucency around the humeral stem (Figure 1). Initial laboratory tests revealed a normal erythrocyte sedimentation rate, C-reactive protein and white blood cell count. A shoulder arthrocentesis was attempted by fluoroscopically guided aspiration, but no fluid was obtained.

What is your diagnosis?

See answer on next page.

anteroposterior radiograph of the left shoulder
Figure 1. An anteroposterior (AP) radiograph of the left shoulder shows evidence of humeral component loosening (a). An axial CT scan shows anterior dislocation of the humeral component (b).

Sources: Frank A. Petrigliano, MD, and Evan E. Vellios, MD


Failed TSA secondary to P. acnes infection with severe glenoid bone loss

Despite negative serum inflammatory markers, the patient’s complaint of progressive pain, skin color changes and radiographic evidence of glenoid and humeral component loosening were concerning for infection. The initial stage of two-stage revision arthroplasty was performed demonstrating grossly loose humeral and glenoid components, a large cavitary glenoid bone defect and diffuse hypertrophic capsular tissue.

Multiple intraoperative cultures were obtained, and a hemiarthroplasty spacer impregnated with vancomycin and gentamycin was inserted (Figure 2). Multiple intraoperative cultures grew Propionibacterium acnes on postoperative day 6 and the patient was placed on the appropriate IV antibiotic therapy (vancomycin) for a total of 6 weeks.


At the end of the initial 6-week IV antibiotic course, the patient was noted to have developed significant pain, erythema and swelling around the surgical site. Inflammatory markers were within normal limits and the patient was afebrile. Due to the concern for ongoing infection, the patient underwent repeat irrigation and debridement (I&D) of the left shoulder and antibiotic-impregnated hemiarthroplasty exchange was performed. Exudate and numerous small pockets of serosanguinous fluid were found in the subcutaneous tissues, as well as in the humeral canal. Intraoperative cultures were again taken, and the patient was placed back on the appropriate IV antibiotic therapy. All repeat cultures were negative. Following an additional 6-week course of IV antibiotic therapy, the patient noted a complete resolution of his prior shoulder pain and his inflammatory markers remained within normal limits.

glenoid bone loss
Figure 2. Shown post-insertion is the status of the antibiotic-impregnated hemiarthroplasty spacer axial (a) and a 3-D CT reconstruction that shows the massive glenoid bone loss (b).
Figure 3. Exposure and preparation of the proximal humerus are shown.
Figure 4. The glenoid surgical exposure and preparation are shown.

The decision was then made to take the patient to the OR for a final left-shoulder I&D, removal of the antibiotic spacer and conversion to left reverse TSA with a custom glenoid base plate to compensate for his glenoid bone loss. Prior to the OR, a patient-specific glenoid was created using the Comprehensive Vault Reconstruction System or VRS (Zimmer Biomet), which is glenoid vault reconstruction software. The implant is designed to restore the native glenoid version and offset and provide screw trajectories that engage the most adequate bone of the scapula available.

The patient received an interscalene block and underwent general anesthesia. He was placed in the beach chair position and an incision was made over the deltopectoral interval. Sharp dissection was used to expose the entire proximal humeral shaft to the deltoid insertion.

The shoulder was dislocated, and the proximal shaft was recut in 20° retroversion, and the canal was debrided with a back-facing curette and canal brush. Sequential irrigation with betadine, peroxide and bacitracin was performed. The canal was sequentially reamed and broached to the appropriate size (Figure 3). A circumferential scar resection was undertaken. Extensive and meticulous dissection was required to expose the glenoid, at which time significant bone loss was again noted (Figure 4).

glenoid baseplate insertion
Figure 5. The custom glenoid baseplate is shown prior to insertion (A), during provisional fixation (B), at final fixation (C) and with the attached glenosphere (D).
Figure 6. Shown are the final humeral stem and bearing tray.

Baseplate trial, placement

The glenoid baseplate trial was placed on the glenoid and appropriate apposition to the remnant glenoid was noted. This was confirmed on an external 3-D-printed model. The actual custom baseplate was then fixed on the glenoid with 2.7 K-wires. The center drill hole was made. The center 6.5-mm screw measured 40 mm and had excellent purchase. The remaining 4.75-mm screws were inserted in standard fashion, and the baseplate was secured (Figure 5).


The glenosphere was then trialed and it adequately covered the inferior aspect of the glenoid. The final glenosphere was inserted, which adequately covered the glenoid. The stem was then trialed with a standard humeral bearing and tray. It could be reduced and had appropriate deltoid tension. The canal was lavaged copiously, and then the final stem was impacted.

The final humeral bearing was inserted, and the shoulder was found to be stable throughout a ROM (Figure 6). A drain was placed exiting posterolaterally. The wound was copiously irrigated, a gram of vancomycin powder was dispersed in the wound, and then the deltopectoral interval was closed in layers.

The patient was placed in a shoulder immobilizer for 6 weeks postoperatively. At 2 weeks postoperatively, the patient began supervised physical therapy with passive FE to 100° and ER to neutral. The patient was seen again at 6 weeks, 3 months and 7 months postoperatively during which time there was gradual progression in his rehabilitation. Postoperative imaging showed no evidence of hardware complication, dislocation or fracture (Figure 7). At the patient’s 7-month postoperative visit he was noted to have a painless active ROM of 140° FE, 30° ER, and IR to L5.

postoperative AP radiograph
Figure 7. An immediate postoperative AP radiograph shows left reverse TSA with the VRS (a). The AP (b) and lateral (c) images at 7 months postoperatively show no evidence of component loosening or fracture.


Complex glenoid bone deformity and deficiency is a growing problem in revision TSA. Severe glenoid bone loss leads to medialization of the glenohumeral joint line and inadequate bony fixation for arthroplasty implants. Current methods to address this include glenoid bone grafting, specialized glenoid implants with posterior augmentation and patient-specific customized glenoid implants created using 3-D CT scan reconstructions.

Scalise and colleagues have shown glenoid bone grafting in the revision setting has a high rate of bone graft subsidence and failure. Similarly, Wagner and colleagues showed patients who undergo glenoid bone grafting experienced a higher rate of reoperation despite subjective improvement in pain. Furthermore, bone grafting in the revision setting requires the use of either bone allograft, which may have an increased rate of failure compared to autograft, or iliac crest autograft with its associated donor site morbidity.

Cil and colleagues showed non-patient-specific customized glenoid components have a high rate of loosening and component subluxation, which requires revision surgery. However, Sandow and colleagues and Schutz and colleagues showed correction of up to 25° retroversion with good osteointegration of the components at 2-year follow-up in patients undergoing primary TSA. These promising results have not been shown in the revision setting or in cases of prior periprosthetic infection.


Custom glenoid bases as described by Dines and colleagues can be promising for dealing with massive glenoid bone loss in primary or revision reverse TSA because they allow for metal augmentation in cases of deficient bone stock and provide custom-placed screw holes directed toward areas with preserved bone stock. Although little research exists on this technique in relation to shoulder arthroplasty, Berasi and colleagues showed this concept can lead to good functional and radiographic outcomes up to 2 years postoperatively in patients with massive acetabular bone loss who undergo revision total hip arthroplasty. As the U.S. population continues to age and more patients undergo shoulder arthroplasty, it is important to recognize this technique as a viable option for managing severe glenoid defects.

Disclosures: Petrigliano reports he is a paid presenter or speaker for Zimmer Biomet and Stryker. Vellios reports no relevant financial disclosures.