The ulna often is considered as a straight bone, around which the radius winds during pro- and supination. However, the anatomy of the ulna is much more complex. The ulna has cartilage coverage in three distinct parts, as it is involved in three different articulations. Proximally there is the semilunar notch, a large depression formed by the olecranon and coronoid process, which articulates to the trochlear part of the distal humerus. Laterally on the coronoid process there is a narrow, oblong, articular depression called the radial notch. It receives the circumferential articular surface of the head of the radius. The head of the ulna is an articular, rounded structure, which articulates laterally to the ulnar notch of the distal radius and distally to the triangular fibrocartilagineous complex. Posteromedial to the ulnar head, the styloid process is a small non-articular eminence.

In the proximal part of the ulna, three bends can be identified. According to Puchwein et al. [13] the mean dorsal hook angle measures 95.3° ± 9.0° with a mean distance from the tip of the olecranon to the edge point of 24.7 ± 2.7 mm. The authors describe a mean varus angulation of 14.3° ± 3.6° and a mean anterior angulation of 6.2° ± 2.7°. The mean varus angulation point and mean anterior angulation points were measured at 75.0 ± 7.9 mm and 44.4 ± 7.4 mm respectively. Grechening et al. [14] describe a varus angulation of 17.5° and an anterior angulation of 4.5°. Windisch et al.[15] record a varus angulation of 17.7° and a mean apex point at 85.4 mm.

Based upon these data, each nail design has to take these angulations into account. A straight nail will not reduce the anatomical axes of the proximal ulna.

The shape of the medullary canal does not correlate with the shape of the posterior border of the ulna. Windisch et al. [16] did demonstrate that the point of varus angulation of the posterior border and the point of varus angulation of the medullary cavity differs from 1 to 60 mm in the majority of specimen. The point of varus angulation of the posterior border was found to be more distal in most cases. Distal to this point of angulation, the medullary canal has a straight trajectory. The varus angle of the medullary cavity ranges from 4° to 13.5° (mean 8.95°), the diameter of the medullary cavity from 4 to 10.5 mm (mean 6.96 mm).

Keener et al. [17] analyzed the anatomy of the insertion of the triceps tendon, as they recognized that triceps splitting approaches can result in insufficiency of the extensor mechanism. They described a distinct lateral tendon expansion continuous with the anconeus fascia (mean width, 16.8 mm). The mean width of the proper triceps tendon was 23.7 mm. The mean maximum olecranon width was 26.9 mm. The ratio of the triceps tendon width to the olecranon width averaged 0.88. The mean thickness of the central tendon insertion was 6.8 mm. The medial part of the triceps tendon showed a distinct, rolled medial edge and an insertion consistently confluent with the central tendon. The triceps footprint insertion was dome shaped. The mean insertional width and length of the tendon was 20.9 and 13.4 mm respectively. The mean distance from the olecranon tip to the tendon was 14.8 mm. The tendon width, thickness, and insertional dimensions correlated with the olecranon width. Based upon their observations, the authors advocate splitting the central tendon 3–5 mm lateral to the midline. This would provide a larger medial tendon flap for secure triceps repair while not compromising the wider, reinforced lateral triceps tendon.

The patient can be installed prone, supine or in a lateral decubitus position. It is important to be able to flex the elbow at 90° and to get a true anteroposterior and lateral fluoroscopic view. After sterile draping, the fracture is reduced either in a closed or in an open way. Reduction is held with K-wires or reduction clamps. If K-wires are used it is important that their position does not interfere with the trajectory of the nail.