There is little information available at present regarding the mechanisms of failure of modern metallic radial head implants. Between 1998 and 2008, 44 consecutive patients (47 elbows) underwent removal of a failed metallic radial head replacement. In 13 patients (13 elbows) the initial operation had been undertaken within one week of a fracture of the radial head, at one to six weeks in seven patients (seven elbows) and more than six weeks (mean of 2.5 years (2 to 65 months)) in 22 patients (25 elbows). In the remaining two elbows the replacement was inserted for non-traumatic reasons. The most common indication for further surgery was painful loosening (31 elbows). Revision was undertaken for stiffness in 18 elbows, instability in nine, and deep infection in two. There were signs of over-lengthening of the radius in 11 elbows. Degenerative changes were found in all but one. Only three loose implants had been fixed with cement. Instability was not identified in any of the bipolar implants.

We undertook this study to determine the minimum amount of coronoid necessary to stabilise an otherwise intact elbow joint. Regan–Morrey types II and III, plus medial and lateral oblique coronoid fractures, collectively termed type IV fractures, were simulated in nine fresh cadavers. An electromagnetic tracking system defined the three-dimensional stability of the ulna relative to the humerus. The coronoid surface area accounts for 59% of the anterior articulation. Alteration in valgus, internal and external rotation occurred only with a type III coronoid fracture, accounting for 68% of the coronoid and 40% of the entire articular surface. A type II fracture removed 42% of the coronoid articulation and 25% of the entire articular surface but was associated with valgus and external rotational changes only when the radial head was removed, thereby removing 67% of the articular surface.

We conclude that all type III fractures, as defined here, are unstable, even with intact ligaments and a radial head. However, a type II deficiency is stable unless the radial head is removed. Our study suggests that isolated medial-oblique or lateral-oblique fractures, and even a type II fracture with intact ligaments and a functional radial head, can be clinically stable, which is consistent with clinical observation.

The lateral ligament complex is the primary constraint to posterolateral rotatory laxity of the elbow, and if it is disrupted during surgery, posterolateral instability may ensue. The Wrightington approach to the head of the radius involves osteotomising the ulnar insertion of this ligament, rather than incising through it as in the classic posterolateral (Kocher) approach. In this biomechanical study of 17 human cadaver elbows, we demonstrate that the surgical approach to the head can influence posterolateral laxity, with the Wrightington approach producing less posterolateral rotatory laxity than the posterolateral approach.