Acute open reduction and internal fixation versus nonoperative treatment of scapular spine fractures after reverse shoulder arthroplasty: a propensity score-matched study

Abstract

Background

Scapular spine fractures occur in 1%-11% of cases after reverse shoulder arthroplasty (RSA). While nonoperative management has been recommended, poor clinical outcomes have been reported for Levy type II-III fractures. The literature on surgical management mostly includes open reduction and internal fixation (ORIF) for established nonunions with mixed results. Currently, little evidence exists on acute ORIF of displaced scapular spine fractures after RSA. Therefore, the purpose of this propensity score (PS)-matched study was to compare acute ORIF (<6 weeks) with nonoperative management.

Methods

This retrospective case-control study included 74 Levy type II-III fractures from 3 surgical centers. Sixteen patients underwent acute ORIF within 6 weeks of symptom onset, and 58 patients underwent nonoperative management. In the operative group, 11 patients (69%) were treated with 90:90° double plating, 4 with hook plating (25%), and 1 (6%) with a hybrid double construct. Clinical (Subjective Shoulder Value, American Shoulder and Elbow Surgeons score, Visual Analog Scale, patient satisfaction, and active forward elevation) and radiological outcomes were compared. PS matching was performed using 5 criteria (age, sex, body mass index, diagnosis, and Levy type) to balance the covariates and to reduce selection bias.

Results

Of the total group (74 fractures: 16 ORIF and 58 nonoperative), 14 patients in each group were matched using the PS. At a mean follow-up of 3.7 years, the mean Subjective Shoulder Value was significantly better (P = .005) for the ORIF group (90%, interquartile range [IQR] 76-90) compared to the nonoperative group (57%, IQR, 42-70). The mean American Shoulder and Elbow Surgeons score (82, IQR 74-90 vs. 59, IQR 40-70), and mean active forward elevation (123°, IQR 95°-137° vs. 75°, IQR 50°-100°) were significantly better in the ORIF group (P = .007 and P = .021, respectively) compared to the nonoperative group. The union rate for the ORIF group was 93% (13/14), compared to 30% (4/13, 1 undetermined) for nonoperative group (risk ratio 0.13, P = .001).

Conclusion

Acute ORIF for displaced Levy-type II and III scapular spine fractures after RSA is associated with significantly better clinical outcomes and union rates compared to nonoperative treatment.