Fractures of the humeral diaphysis occur in a bimodal distribution and represent 3-5% of all fractures. Presently, the standard treatment of isolated humeral diaphyseal fractures is nonoperative care using splints, braces, and slings. Recent data has questioned the effectiveness of this strategy in ensuring fracture healing and optimal patient function. The primary objective of this randomized controlled trial (RCT) was to assess whether operative treatment of humeral shaft fractures with a plate and screw construct provides a better functional outcome than nonoperative treatment. Secondary objectives compared union rates and both clinical and patient-reported outcomes. Eligible patients with an isolated, closed humeral diaphyseal fracture were randomized to either nonoperative care (initial sugar-tong splint, followed by functional coaptation brace) or open reduction and internal fixation (ORIF; plate and screw construct). The primary outcome measure was the Disability Shoulder, Arm, Hand (DASH) score assessed at 2-, 6-, 16-, 24-, and 52-weeks. Secondary outcomes included the Short Musculoskeletal Functional Assessment (SMFA), the Constant Shoulder Score, range of motion (ROM), and radiographic parameters. Independent samples t-tests and Chi-squared analyses were used to compare treatment groups. The DASH, SMFA, and Constant Score were modelled over time using a multiple variable mixed effects model. A total of 180 patients were randomized, with 168 included in the final analysis. There were 84 patients treated nonoperatively and 84 treated with ORIF. There was no significant difference between the two treatment groups for age (mean = 45.4 years, SD 16.5 for nonoperative group and 41.7, SD 17.2 years for ORIF group; p=0.16), sex (38.1% female in nonoperative group and 39.3% female in ORIF group; p=0.87), body mass index (mean = 27.8, SD 8.7 for nonoperative group and 27.2, SD 6.2 for ORIF group; p=0.64), or smoking status (p=0.74). There was a significant improvement in the DASH scores at 6 weeks in the ORIF group compared to the nonoperative group (mean=33.8, SD 21.2 in the ORIF group vs. mean=56.5, SD=21.1 in the nonoperative group; p < 0 .0001). At 4 months, the DASH scores were also significantly better in the ORIF group (mean=21.6, SD=19.7 in the ORIF group vs. mean=31.6, SD=24.6 in the nonoperative group; p=0.009. However, there was no difference in DASH scores at 12-month follow-up between the groups (mean=8.8,SD=10.9 vs. mean=11.0, SD=16.9 in the nonoperative group; p=0.39). Males had improved DASH scores at all timepoints compared with females. There was significantly quicker time to union (p=0.016) and improved position (p < 0 .001) in the ORIF group. There were 13 (15.5%) nonunions in the nonoperative group and four (4.7%) combined superficial and deep infections in the ORIF group. There were seven radial nerve palsies in the nonoperative group and five (a single iatrogenic) radial nerve palsies in the ORIF group. This large RCT comparing operative and nonoperative treatment of humeral diaphyseal fractures found significantly improved functional outcome scores in patients treated surgically at 6 weeks and 4 months. However, the early functional improvement did not persist at the 12-month follow-up. There was a 15.5% nonunion rate, which required surgical intervention, in the nonoperative group and a similar radial nerve palsy rate between groups
