Session VIII - Upper Extremity

A Randomized Prospective Trial of Humeral Shaft Fracture Fixation: Compression Plate vs. Intramedullary Nail

R.G. McCormack, MD, FRCS (C); D. Brien; Richard E. Buckley, MD, FRCS (C); Michael D. McKee, MD, FRCS (C); James N. Powell, MD, FRCS (C); Emil H. Schemitsch, MD, FRCS (C); Royal Columbian Hospital, New Westminster, BC; Foothills Medical Center, Calgary, AB; St. Michael's Hospital, Toronto, ON

Objective: To compare functional outcomes of locked intramedullary nailing to plate and screw fixation of humeral shaft fractures.

Design: Multi-centre prospective randomized clinical trial.

Setting: Three Canadian teaching hospitals.

Patients: Skeletally mature patients with nonpathologic fracture of the humeral shaft and indications for surgical stabilization. Twenty-four patients were randomized to the plate group and twenty-one were randomized to the nail group. Two patients died and one patient was lost to follow-up.

Intervention: Preoperative randomization to undergo open reduction and internal fixation with dynamic compression plate (DCP) or implantation of a locked intramedullary nail (IM).

Main Outcome Measures: The primary outcomes were function, as assessed by the Society of American Shoulder and Elbow Surgeons (ASES) shoulder scores for ADL's and visual analogue scores for pain. The secondary outcomes were the incidence of complications and the need for subsequent surgery.

Results: Forty-one patients (IM:19, DCP:22) had a minimum of six months followup. There were similar demographics, fracture types and ISS scores for the two groups. There was no significant difference in ASES scores (IM:47/52, DCP:48/52), VAS pain scores (IM:1.5/10, DCP:1/10), strength, ROM or return to activity. Despite this, shoulder impingement symptoms were more common in the IM group (IM:6, DCP:1), and five of the six cases of impingement in the IM group occured after antegrade nail insertion. There was no significant difference in OR time or blood loss. The DCP group was more likely to be braced postoperatively (8/23 vs 2/21) and have a bone graft performed (4/23 vs 0/21). The IM group had more complications (IM:13, DCP:3) and significantly more secondary surgery (IM:7, DCP:1) with P = 0.016. There were three transient radial nerve injuries in the IM group, vs none in the DCP.

Conclusion: Functional outcomes, at minimum six-month follow-up, are similar between the two groups as assessed by ASES shoulder scores and VAS scores for pain. Antegrade nail insertion is more likely than retrograde nail or plate fixation to cause impingement symptoms. ORIF with a DC plate remains the gold standard for surgical intervention of humeral diaphyseal fracture. IM stabilization may be indicated for specific situations, but insertion is technically more demanding, has a higher complication rate and a higher reoperation rate (P = 0.016).