Some in the medical community have expressed concern about performing total shoulder arthroplasty (TSA) for instability arthropathy in patients who have undergone a prior coracoid transfer procedure due to the distorted anatomy, scarring, and concerns about early failure or poor outcomes. Some have advocated for reverse TSA in all such patients; however, researchers recently compared outcomes for patients who underwent TSA after coracoid transfer versus patients who had TSA for primary osteoarthritis (OA) and found no significant differences in outcomes.

Michael J. Bender, MD, an orthopaedic shoulder and elbow surgeon at Methodist Sports Medicine in Indianapolis, presented the data during the American Shoulder and Elbow Surgeons (ASES) Specialty Day at the AAOS 2019 Annual Meeting.

“Our biggest takeaway is that we don’t think anatomic TSA should be abandoned yet as a valuable option for these typically young and challenging patients,” Dr. Bender said.

The researchers conducted a retrospective review from a single-surgeon arthroplasty registry from December 2004 to February 2018. Of the nearly 3,000 patients in the database, 18 had a prior coracoid transfer, and 14 of them received anatomic TSA. Four cases required reverse TSA due to either severe deformity, severe glenoid bone loss, or cuff insufficiency. Eleven patients with a minimum of two years of follow-up were then matched based on age, sex, body mass index, shoulder dominance, and comorbidities to three control subjects who underwent anatomic TSA for primary OA (n = 33).

“Both groups were fairly young, which is common in most instability arthropathy cohorts,” Dr. Bender said.

The average age was 56 years in both cohorts, with an average follow-up of 4.8 and 3.9 years, respectively.

Three patients had a prior Latarjet, eight had a Bristow, and many had soft-tissue procedures prior to coracoid transfer. The average number of years from coracoid transfer to arthroplasty was just less than 29. There was no statistical difference in glenoid wear patterns between the cohorts. Most of the coracoid transfers had no fatty infiltration changes within the subscapularis.

At final follow-up, both cohorts had significant improvement in ASES and Single Assessment Numerical Evaluation (SANE) scores, but there was no statistically significant difference between the groups. Total ASES scores improved from 43.1 preoperatively to 88.3 postoperatively in coracoid transfer patients and from 36.3 to 82.2 in primary OA patients (P < 0.001 for both).

There was no statistically significant difference in ASES final scores (88.3 versus 82.2; P = 0.166) or overall improvement in scores (P = 0.954) between the groups. ASES pain scores improved from 20.0 to 45.0 in the coracoid transfer cohort and from 8.4 to 40.6 in the primary OA cohort (P < 0.001 for both). ASES pain score was greater in the coracoid transfer group (45.0 versus 40.6; P = 0.004), but the improvement in pain scores was similar (P = 0.183). SANE scores improved from 31.6 to 84.6 in coracoid transfer patients and from 28.2 to 66.8 in primary OA patients (P < 0.001 for both). Neither the final SANE score (84.6 versus 66.8; P = 0.120) nor improvement in the score (P = 0.293) differed significantly between groups.

“We believe this represents equivalent results to TSA for primary OA at this follow-up interval,” Dr. Bender said.

Both cohorts reported good overall satisfaction at follow-up: 90 percent of patients in the coracoid group and 81.8 percent in the primary OA group reported being “satisfied” or “very satisfied” (P = 0.784).

Complications occurred in three patients (n = 3 complications) in the coracoid group and four patients (n = 7 complications) in the primary OA group. The complications in the coracoid group were stitch abscess, deep infection, and aseptic glenoid loosening/polyethylene wear (n = 1 for each). The complications in the primary OA cohort were stitch abscess (n = 1), neuropraxic injuries (n = 2), small glenoid fracture (n = 1), aseptic glenoid loosening (n = 2), and deep infection (n = 1).

“Interestingly, there were no nerve injuries in this cohort, which has been a previous concern with the distorted anatomy,” said Dr. Bender.

In each group, two patients required revisions due to deep infection or glenoid loosening.

The study is limited by its retrospective design, small patient population, and use of a single-surgeon registry. There was also insufficient postoperative radiographic data to include in the analysis.

“The mid-term follow-up may not be long enough to detect true differences in revision rates over time in a young patient cohort for which longevity is important, but early results are encouraging,” Dr. Bender noted.

Dr. Bender’s coauthors of “Outcomes of Total Shoulder Arthroplasty for Instability Arthropathy with a Prior Coracoid Transfer Procedure: A Retrospective Review and Comparative Cohort” are Brent J. Morris, MD; Mitzi S. Laughlin, PhD; Aydin Budeyri, MD; Ryan K. Le, BS; Hussein A. Elkousy, MD; and T. Bradley Edwards, MD.

Kerri Fitzgerald is the managing editor of AAOS Now. She can be reached at kefitzgerald@aaos.org.