Question: In older patients with displaced, comminuted intra-articular distal humeral fractures, is primary total elbow arthroplasty more effective than open reduction and internal fixation in improving functional outcomes and reducing reoperation rates?
Design: Randomized (allocation concealed)*, blinded (outcome assessor) controlled trial with 24-month follow-up.
Setting: 4 university-affiliated medical centers in Canada.
Patients: 42 patients >65 years of age with displaced, comminuted, intra-articular fractures of the distal part of the humerus requiring surgery. Closed fractures, Gustilo grade-I open fractures treated within 12 hours of injury, and Gustilo grade-I open fractures treated elsewhere by irrigation and débridement within 12 hours after injury were included. Exclusion criteria included extra-articular or partial articular fractures; intra-articular fractures not requiring surgery; Gustilo grade-I fractures that had not had irrigation and débridement within 12 hours; Gustilo grade-II, IIIa, IIIb, and IIIc open fractures; associated vascular injury; previous ipsilateral distal humeral fracture; pathologic fractures; fractures with diaphyseal extension =8 cm; definitive surgery >21 days after injury; preexisting severe joint disease; dementia; and substantial comorbid conditions. 40 patients (95%) (mean age, 78 y; 88% women) completed the study.
Intervention: Patients were allocated to receive total elbow arthroplasty (n = 20) or open reduction and internal fixation (n = 20). For total elbow arthroplasty, either a midline triceps-splitting or a triceps-sparing approach was used, allowing the surgeon to use the working space created by the condylar resection to do the procedure without detaching the triceps from the olecranon. Open reduction and internal fixation was done through a posterior approach and with use of a midline triceps-splitting incision or an olecranon osteotomy with anatomic reduction of the fracture and provisional Kirschner-wire fixation. Definitive fixation was done with 2 small-fragment compression plates. Intraoperative conversion to total elbow arthroplasty in the open reduction and internal fixation group was anticipated if the surgeon judged that sufficient stability to allow early mobilization could not be obtained with open reduction and internal fixation. Patients in both groups followed the same rehabilitation protocol.
Main outcome measures: The primary outcome was reoperation rate. Secondary outcomes were patient outcome measures (pain, motion, stability, and daily function) based on the Mayo Elbow Performance Score (MEPS) and the Disabilities of the Arm, Shoulder and Hand (DASH) scale. Complication rates were also measured.
Main results: Intention-to-treat and on-treatment analyses were both performed. 5 patients underwent conversion from open reduction and internal fixation to total elbow arthroplasty during surgery. The study had 80% power to detect a 40% difference in reoperation rate. The groups did not differ significantly with regard to reoperation rate (Table). The mean MEPS was improved in patients who received a total elbow arthroplasty compared with those who received open reduction and internal fixation at all follow-up time points (Table). DASH scores were better in the total elbow arthroplasty group at 6 months, but the group scores were not significantly different at 12 and 24 months (Table). Complication rates were similar in the total elbow arthroplasty and open reduction and internal fixation groups (10 vs 8 patients had =1 complication, p = 0.4).
Conclusions: In older patients with displaced, comminuted intra-articular distal humeral fractures, primary total elbow arthroplasty was not more effective than open reduction-internal fixation in reducing reoperation rates. Total elbow arthroplasty was associated with improved functional outcomes based on the Mayo Elbow Performance Score.
(*Information provided by author.)
The study by McKee and colleagues is important because elbow replacement is increasingly being performed for patients with severely comminuted fractures at the elbow. Several reports have documented the value of elbow replacement in the setting of a comminuted distal humeral fracture in the older individual. Because this fracture is not common, a multisite enrollment protocol is acceptable and even essential under these circumstances.
The central question concerns whether older patients with comminuted distal humeral fractures exhibit the best functional outcome in the first two years after open reduction and internal fixation or total elbow replacement. This question was immediately complicated by the observation that 20% of the fractures initially allocated to open reduction and internal fixation technically could not be adequately fixed in the judgment of the surgeon and, hence, these patients were moved to the replacement group. While the results were analyzed as intention to treat, it is important to recognize these crossovers. Nonetheless, the MEPS improved for those with joint replacement at all time periods up to 2 years after surgery. Although the DASH score also favored replacement in the early period, the difference was not significant at the time of the 2-year follow-up.
One area of concern was the power statement indicating that a 40% difference in outcome was needed to reach 80% confidence that a difference does exist. The small sample size does not allow demonstration of significance, as is expressed by the authors. Yet, I believe that this conclusion, while technically accurate, is potentially misleading. From a practical perspective, most orthopaedic surgeons would consider a reoperation rate of 12% in the arthroplasty group to be meaningfully different from a 27% reoperation rate in the open reduction and internal fixation group. If one event occurs twice as often as the other, although not significant, it might be considered relevant from a clinical perspective. Otherwise, the conclusions are valid and supported by the analyses. An important observation was that the outcome expected after total elbow arthroplasty was probably optimum at 1 year following the implantation. Complications such as wear and loosening occur only with longer surveillance. On the other hand, typically, the open reduction and internal fixation group would be considered stable over the course of time.
This study by McKee and colleagues is a valuable contribution to the orthopaedic literature and provides a scientific basis for clinical practice and decision-making.