Operative versus nonoperative treatment of proximal humeral fractures : a systematic review, meta-analysis, and comparison of observational studies and randomized controlled trials

Background: There is no consensus on the choice of treatment for displaced proximal humeral fractures in older patients (aged > 65 years). The aims of this systematic review and meta-analysis were (1) to compare operative with nonoperative management of displaced proximal humeral fractures and (2) to compare effect estimates obtained from randomized controlled trials (RCTs) and observational studies. Methods: The databases of MEDLINE, Embase, CENTRAL (Cochrane Central Register of Controlled Trials), and CINAHL (Cumulative Index to Nursing and Allied Health Literature) were searched on September 5, 2017, for studies comparing operative versus nonoperative treatment of proximal humeral fractures; both RCTs and observational studies were included. The criteria of the Methodological Index for Non-Randomized Studies, a validated instrument for methodologic quality assessment, were used to assess study quality. The primary outcome measure was physical function as measured by the absolute Constant-Murley score after operative or nonoperative treatment. Secondary outcome measures were major reinterventions, nonunion, and avascular necrosis. Results: We included 22 studies, comprising 7 RCTs and 15 observational studies, resulting in 1743 patients in total: 910 treated operatively and 833 nonoperatively. The average age was 68.3 years, and 75% of patients were women. There was no difference in functional outcome between operative and nonoperative treatment, with a mean difference of –0.87 (95% confidence interval, –5.13 to 3.38; P =.69; I2 = 69%). Major reinterventions occurred more often in the operative group. Pooled effects of RCTs were similar to pooled effects of observational studies for all outcome measures. Conclusions: We recommend nonoperative treatment for the average elderly patient (aged > 65 years) with a displaced proximal humeral fracture. Pooled effects of observational studies were similar to those of RCTs, and including observational studies led to more generalizable conclusions