Improving preoperative planning of revision surgery after previous anatomic total shoulder arthroplasty.

ABSTRACT

BACKGROUND: The purpose of this study was to compare preoperative radiographic evaluation with intraoperative video and explant analysis in patients undergoing revision of a previous anatomic total shoulder arthroplasty (TSA). METHODS: We evaluated the preoperative radiographs of 165 revisions of failed TSAs for component loosening and glenohumeral registry (ie, the spatial relationship of the glenoid component and the prosthetic humeral head). Seventy-nine intraoperative videos were evaluated for component stability, rotator cuff (RC) integrity, synovitis, and glenoid bone loss. Eighty-seven explants were reviewed to assess wear patterns and presence of backside cement. RESULTS: Of 79 glenoid components, 47 were radiographically loose, but only 30 of 79 were loose intraoperatively. Thirty-two were radiographically fixed, but only 26 of 32 were fixed intraoperatively. If radiographically loose, 53% had severe glenoid bone loss. If radiographically fixed, 77% had mild to moderate bone loss (P = .008). Synovitis was associated with glenoid fixation mild with a loose glenoid (6%) and severe with a fixed glenoid (30%, P = .012). Superior registry comprised 46%. RC deficiency was associated with posterior and anterior registry (88% and 79%, respectively). Explant examination revealed an eccentric wear pattern was predominant. CONCLUSION: Radiographic evaluation of glenoid loosening in patients undergoing revision of TSAs will often differ from intraoperative findings (40% false-positive rate and 17% false-negative rate). Assessment of glenohumeral registry can help anticipate RC deficiency, with posterior and anterior registry associated with RC deficiency. Patients with a loose glenoid are more likely to have severe synovitis and more severe glenoid bone deficiencies. Failed TSAs are more likely to have asymmetrical wear of the glenoid component, suggesting altered pathomechanics that may have led to failure.