Routine use of autograft is not necessary for treatment of humeral shaft nonunions and anticipated nonunions after failed nonoperative treatment

Objectives: To determine if the use of autogenous bone graft is necessary for the treatment of humeral shaft nonunions after initial nonoperative management. Design: Retrospective Setting: Level 1 academic trauma center Patient Selection Criteria: All skeletally mature patients undergoing nonunion repair of a humeral shaft fracture (AO/OTA 12A, 12B, 12C) after initial nonoperative management of the acute fracture. Outcome Measures and Comparisons: The primary outcome was osseous union. Failure of nonunion repair was defined by lack of osseous union within 365 days from surgery and/or return to the operating room for additional attempts to promote union. Secondary outcomes included complications including infection, radial nerve palsy, and donor site morbidity. Results: Seventy-two patients were included in the final cohort. Thirty-eight patients (53%) were female, and the average age was 51 (SD 18, range 17, 83). Two patients (3%) developed a recalcitrant nonunion, both of which healed after a second procedure consisting of revision compression plating. The use of bone autograft was uncommon overall in this cohort (4 patients, 6%). Bone morphogenic protein, bone allograft, or demineralized bone matrix was used in 12 patients (17%). The remaining 56 patients (78%) were treated with compression plating alone. Compression was generated through multiple techniques including use of the articulated tensioning device, a pull screw with a verbrugge clamp, lag screw application, and compression generated via eccentric drilling through the plate. There were no differences in terms of patient demographics, fracture or injury characteristics between the groups that received autograft and those who did not (p>0.05). Use of autograft or other biologic supplementation was not associated with a statistically significant increase in union rate, 100% versus 97% (p=1.00). In the four patients who underwent autogenous bone grafting, there were no reported donor site complications. Conclusions: For humeral shaft fractures initially treated nonoperatively that went on to nonunion or anticipated nonunion, the union rate for compression plating alone was comparable to the union rate noted in the literature after treatment with bone autograft and compression plating. These results suggest that routine use of bone autograft may be unnecessary in the treatment of humeral shaft nonunions.