Radiographic loss of contact between radial head fracture fragments is moderately reliable

Abstract

Background: Loss of contact between radial head fracture fragments is strongly associated with other elbow or forearm injuries. If this finding has adequate interobserver reliability, it could help examiners identify and treat associated ligament injuries and fractures (eg, forearm interosseous ligament injury or elbow dislocation).

Questions/purposes: (1) What is the interobserver agreement on radiographic loss of contact between radial head fracture fragments? (2) Are there factors associated with the observer such as location of practice or subspecialization that increase interobserver reliability?

Methods: Fully trained practicing orthopaedic and trauma surgeons from around the world evaluated 27 anteroposterior and lateral radiographs of radial head fractures on a web-based platform for the following characteristics: (1) loss of contact between at least one radial head fracture fragment and the remaining radial head and neck; (2) a gap between fragments of 2 mm or greater; (3) anticipated fracture instability (mobility) on operative exposure; (4) anticipated associated ligament injuries; and (5) recommendation for treatment. Agreement among observers was measured using the multirater kappa measure. Kappas for various observer characteristics were compared using 95% confidence intervals.

Results: The overall interobserver agreement was moderate (range, 0.49-0.55) for each question except associated ligament injury, which was fair (0.33). Shoulder and elbow surgeons had substantial agreement (range, 0.51-0.61) in many areas, but kappas were generally in the moderate range (0.41-0.59) based on number of years in practice, radial head fractures treated per year, and trainee supervision.

Conclusions: Radiographic signs of radial head fracture instability such as loss of contact have moderate reliability. This characteristic seems clinically useful, because loss of contact between at least one radial head fracture fragment and the remaining radial head and neck is strongly associated with associated ligament injury or other fractures.

Level of evidence: Level III, diagnostic study. See Guidelines for Authors for a complete description of levels of evidence.

Fig. 1

AP radiograph showing a displaced whole head radial head fracture.

Fig. 2

Lateral radiograph showing a displaced whole head radial head fracture.

Fig. 3

AP radiograph showing a nondisplaced partial radial head fracture.

Fig. 4

Lateral radiograph showing a nondisplaced partial radial head fracture.