Revisiting the Socket: Commentary on an article by Jeremy S. Somerson, MD, et al.
Prosthetic hemiarthroplasty combined with joint shaping and smoothing to create a fibrocartilage interface is a strategy for treating arthrosis that dates back to the glass and Vitallium mold arthroplasties first described for the hip by Smith-Petersen nearly 80 years ago2,3. Matsen et al. demonstrated in a cadaver model that a concentrically reamed glenoid surface provides glenohumeral stability similar to an all-polyethylene glenoid implant4. They also studied the ream-and-run procedure in a living canine model5 to demonstrate that carefully reamed glenoid bone articulating with a spherical humeral hemiarthroplasty implant “heals and remodels to a concentric, smooth, fibrocartilage articular surface.”6 Their work also established that reaming to a diameter of curvature 2 mm greater than that of the prosthetic humeral head was optimal to balance joint stability and mobility, including humeral head translation.
Matsen et al. have reported extensively on their experience with this procedure6-8. In a case-controlled study comparing the ream-and-run procedure with anatomic total shoulder arthroplasty, Clinton et al. concluded that both procedures offer similar functional recovery although the time to recovery following the ream-and-run procedure may be longer6. Gilmer et al. also found that results continued to improve for 18 months postoperatively, but younger patients and those with previous surgery had an outcome inferior to that of older patients without prior surgery7. This is disconcerting because the ream-and-run procedure was originally conceived as an attractive alternative for younger active patients desiring to maintain high activity levels postoperatively who might otherwise be at risk of early glenoid implant loosening. Thus, an effective durable surgical option for young patients with advanced glenohumeral arthritis, who may have a history of 1 or more operations for instability or resultant chondrolysis, remains elusive. Another study of patients with a retroverted, biconcave glenoid and posterior humeral head subluxation who underwent this procedure without any deliberate attempt to correct retroversion noted improvements in humeral head centering on postoperative radiographs8.
Additionally, there has been little outside study of the ream-and-run procedure. Somerson and Wirth demonstrated that 14 of 17 patients achieved a minimum clinically important improvement in outcome following hemiarthroplasty with concentric glenoid reaming, while 3 patients underwent revision9. Kearns et al. reported only modest improvement with respect to pain and function at short-term follow-up after hemiarthroplasty with concentric glenoid reaming10. In addition, 5 of their 36 patients underwent revision surgery for residual pain within 1 year of surgery. However, their cohort included patients with chondrolysis, dislocation arthropathy, and glenoid dysplasia—i.e., patients who were vastly different from those in the study by Somerson et al. and not typically considered by Matsen for the ream-and-run procedure11.
Somerson et al. found in their study that outcome did not correlate with the amount of glenoid medialization. However, some patients experienced 5 to 10 mm of medialization within a few years, or a rate of medialization greater than that reported in an earlier radiographic study12. Substantial glenoid wear is concerning, irrespective of short-term outcome, because this may affect rotator cuff function and complicate late revision. The authors should continue their radiographic study of the ream-and-run procedure to identify predictors of rapid glenoid medialization.