Acute Scaphoid Fractures

Brent Graham, M.D., MSc, FRCSC
University Health Network/University of Toronto Hand Program
Toronto, ON

The key to treating acute scaphoid fractures is accurate diagnosis. Despite the conventional wisdom on this topic, an informed observer, such as an orthopaedic surgeon, will identify a scaphoid fracture, even when it is completely undisplaced on the initial X-ray. Obviously, diagnosis requires alertness to the possibility of a scaphoid fracture plus a consistent history combined with physical examination findings of tenderness to palpation over the scaphoid itself. The physical examination should include palpation of the scaphoid not only in the anatomic snuffbox but also dorsally and over the scaphoid tubercle. Tenderness in any one of these locations may be an indication of a scaphoid fracture. The overwhelming majority of scaphoid fractures occur between the ages of 15 and 40. Acute scaphoid fractures are rare outside this age range.


Figure 1. CT-scan undisplaced fracture

When an acute scaphoid fracture is identified on plain radiographs, further imaging, such as a CT scan, may also be required. The reason is that treatment for displaced and nondisplaced scaphoid fractures is different and the judgment of clinically relevant displacement on plain radiographs is simply unreliable. When the plain radiographs show that a scaphoid fracture is present and obviously displaced, operative management is indicated. No additional imaging is required. When the fracture appears to be undisplaced, further imaging with CT scanning is the appropriate next step (Figure 1). CT scanning should also be considered when plain radiographs do not demonstrate a fracture but are in conflict with the clinical evaluation suggesting a strong possibility of an acute scaphoid fracture. Under these circumstances an MRI is useful although it does not necessarily provide the same degree of bone detail as CT scanning. Nuclear medicine scanning should be discouraged because it is so nonspecific and has little clinical usefulness. A combination of normal plain radiographs and CT scan essentially rules out a fracture of the scaphoid.

For scaphoid fractures that are displaced on plain radiographs or which appear to be undisplaced on plain radiography but are shown to be displaced on CT scanning, operative stabilization is recommended. As little as 1 mm of displacement in either the coronal or sagittal planes or angulation at the fracture site beyond 70° between the proximal and distal fragments on the lateral view strongly suggests instability. The objective of surgery is to achieve stability of the fracture fragments because instability is what is implied by this degree of displacement. In other words, the concern is not that there will be a malunion of the fracture fragments if this displacement is not addressed but rather that a nonunion is much more likely if the fracture is treated nonoperatively. The best method of internal fixation is unknown but some type of implant that allows immediate protected active range of motion exercises is preferred. However, even if K-wires are used to stabilize the fragments and postoperative immobilization is required, the chances of successful healing are much greater than with immobilization alone.

For acute fractures of the scaphoid that appear to be undisplaced on plain radiography and on CT scanning, treatment with immobilization in a below elbow thumb spica cast is entirely appropriate. It should be continued until the fracture is clinically healed, usually a period of four to six weeks. Because the scaphoid bone is almost completely enclosed by articular cartilage, there will never be callus at the fracture site. In addition, awaiting the radiographic appearance of trabeculae crossing the fracture site before mobilization will almost always mean that the patient will remain immobilized much longer than necessary.

The alternative to nonoperative treatment would be to perform operative stabilization of the undisplaced fracture. While this may result in a slightly more rapid rehabilitation and return to activities, there is no evidence that the rate of nonunion is reduced by this strategy and there will be risks of operative treatment, principally stiffness of the wrist joint. This is a decision to be made by the surgeon and patient but normally operative management would not be required for these patients and the expectation should be that there will be full, uncomplicated union within four to six weeks.