[Efficacy analysis of two-stage reverse total shoulder arthroplasty for treating postoperative deep infection after surgeries for proximal humeral fractures].

OBJECTIVE

To evaluate the clinical results of two-stage reverse total shoulder arthroplasty for treating postoperative deep infection after surgeries for proximal humeral fractures.

METHODS

From January 2013 to December 2014, 8 consecutive patients with postoperative deep infection after surgeries for proximal humeral fractures who were treated with two-stage reverse total shoulder arthroplasty were retrospectively reviewed after the final follow-up. There were 1 man and 7 women with a mean age of (58.5±6.4) years, of whom 3 left shoulders and 5 right shoulders were involved. There were 2 patients with periprosthetic infection after hemiarthroplasty for proximal humeral fractures, and 6 patients with humeral head necrosis as well as implant-associated infection after open reduction internal fixation for proximal humeral fractures with the locking plate. The diagnosis of postoperative deep infection was confirmed by either the preoperative cultures or the intraoperative biopsies during the first-stage surgery. At the first-stage surgery, all the patients underwent a thorough debridement, and then an antibiotic-impregnated bone cement spacer was placed after the removal of prosthesis or locking plate. During the second-stage surgery, the cement spacer was removed, and then a revision shoulder arthroplasty with the reverse shoulder prosthesis was performed in all the patients who were routinely followed up after the second-stage surgery. The visual analogue score (VAS), Constant score and University of California Los Angeles (UCLA) score were employed to evaluate the postoperative shoulder function.

RESULTS

The mean follow-up time was (19.9±8.0) months (range 12 to 35 months). At the end of the follow-up, the median forward elevation [100° (60°, 140°) vs. 25° (0°, 90°), P=0.011], the median external rotation [15° (0°, 50°) vs. 5° (0°, 20°), P=0.048], and the median internal rotation [L4 (buttock, T12) vs. buttock (buttock, L3), P=0.041] were all significantly improved postoperatively. The median Constant score [53.5 (32, 74) vs. 29.0 (10, 57), P=0.012], the median UCLA score [20.5 (9, 26) vs. 9.5 (5, 15), P=0.012], and the median VAS score [1.5 (0, 5) vs. 5.0 (0, 8), P=0.018] were all significantly improved after the surgery. No recurrence of infection, prosthetic loosening or neurovascular injury was noted by the last follow-up.

CONCLUSION

two-stage reverse total shoulder arthroplasty was an effective treatment for the postoperative deep infection after surgeries for proximal humeral fractures. The shoulder function was postoperatively improved to a certain degree.