RESUMO
BACKGROUND: In smaller studies, allograft-prosthetic composite (APC) has been used for proximal humerus bone loss with some success, although with notable complication risk. This systematic review and meta-analysis sought to describe outcomes and complications after proximal humerus APC and how major APC complications are defined in the literature. METHODS: A systematic review was performed per Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed/MEDLINE, EMBASE, Web of Science, and Cochrane were queried for articles on APC for proximal humeral bone loss secondary to tumor, fracture, or failed arthroplasty. Primary outcomes included postoperative range of motion, outcome scores (Musculoskeletal Tumor Society [MSTS], Simple Shoulder Test [SST], American Shoulder and Elbow Surgeons [ASES], Constant, visual analog scale [VAS], and subjective shoulder value [SSV]), and complication incidence. We also described individual study definitions of APC malunion/nonunion, methods of postoperative evaluation, malunion/nonunion rates, allograft fracture/fragmentation rates, and mean union time, when available. Secondarily, we compared hemiarthroplasty and reverse total shoulder arthroplasty. RESULTS: Sixteen articles including 375 shoulders were evaluated (average age: 49 years, follow-up: 54 months). Fifty-seven percent of procedures were performed for tumors, 1% for proximal humerus trauma sequelae, and 42% for revision arthroplasty. Average postoperative forward elevation was 82° (69-94°), abduction 60° (30-90°), and external rotation 23° (17-28°). Average MSTS score was 82% (77%-87%), SST score 5.3 (4.5-6.1), ASES score 64 (54-74), Constant score 44 (38-50), VAS score 2.2 (1.7-2.7), and SSV 51 (45-58). There was a 51% complication rate with an 18% nonallograft surgical complication rate, 26% APC nonunion/malunion/resorption rate, and 10% APC fracture/fragmentation rate. Fifteen percent of nonunited APCs underwent secondary bone grafting; 3% required a new allograft; and overall revision rate was 12%. APC nonunion/malunion was defined in 2 of 16, malunion/nonunion rates in 14 of 16, fracture/fragmentation rates in 6 of 16, and mean union time (7 months) in 4 of 16 studies. CONCLUSION: APC reconstruction of the proximal humerus remains a treatment option, albeit with substantial complication rates. In addition, there is a need for APC literature to report institutional definitions of nonunion/malunion, postoperative evaluation, and time to union for a more standardized evaluation. LEVEL OF EVIDENCE: Level IV; systematic review. See Instructions for Authors for a complete description of levels of evidence.
