Plain Radiography

Plain radiographs remain the gold standard initial imaging technique for suspected acute scaphoid fractures.8 The American College of Radiology has recommended four views for a suspected scaphoid fracture.9 These are posteroanterior (PA), lateral, semipronated oblique, and PA with ulnar deviation ( Fig. 4.1 ). A meta-analysis of several large series has shown that scaphoid fractures are not identified on initial plain radiographs in 16% of cases.2 ^, 3

The “scaphoid fat stripe” is a small linear collection of fat that lies between the radial collateral ligament and the tendon sheaths of the abductor pollicis longus and extensor pollicis brevis, seen on PA and oblique views as a thin lucent line parallel to the radial border of the scaphoid with a slight convexity toward it.10 It was postulated that trauma to the scaphoid resulted in the accumulation of blood and edema, making the fat stripe more pronounced on radiographs. Annamalai and Raby reported a 50% correlation between MR I-confirmed scaphoid fractures and the presence of the scaphoid fat stripe, making this soft-tissue indicator an unreliable sign for the detection of occult fractures.11

In the case of diagnostic uncertainty in a patient with clinical symptoms and signs of scaphoid fracture but with no evidence of fracture on initial plain radiography, the traditional model has been to repeat plain radiographic examination 2 weeks following the traumatic insult. The rationale for this is that resorption of bone at the fracture margins increases the radiolucency relative to the surrounding bone. However, studies have reported that repeat plain radiographs have a poor interobserver reliability (18 to 53%).3 ^– 6 Traditionally, patients with a suspected scaphoid fracture but with negative initial radiographs were treated with 2 weeks of cast immobilization while awaiting repeat clinical examination and radiography. This inevitably resulted in overtreatment of some wrist injuries, inconvenience and time lost from work for the patient, and increased cost to the health service. Other imaging modalities are therefore required to make an accurate diagnosis in the acute situation when plain radiographs do not demonstrate a fracture.

High-Spatial-Resolution Sonography (5–15 MHz)

Ultrasonography is an uncommonly used technique to diagnose suspected scaphoid fractures. Several studies have reported their findings with high-spatial-resolution sonography.12 ^, 13 These studies suggested that scaphoid cortical interruption, an effusion in the radiocarpal joint, and an effusion in the scaphotrapez trapezoidal (STT) joint are diagnostic of a scaphoid fracture ( Fig. 4.2 ). The advantages of high-spatial-resolution sonography as a secondary imaging modality include accessibility, noninvasiveness, safety, low cost, and a short examination time. The uninjured side can also be examined at the same time for comparison. The disadvantages are that it can assess only the dorsal scaphoid waist, possibly missing fractures of the distal radius or other carpal bones. Also, the tubercle area of the scaphoid may appear quite irregular, mimicking cortical disruption and resulting in false-positives. The most important clinical factor is the experience of the sonographer, and the dynamic nature of this investigation makes it difficult for the treating clinician to interpret the stored images. For these reasons it is not commonly used in clinical practice.

Bone Scintigraphy

Bone scintigraphy may be used as a second-line imaging modality when there is clinical suspicion in patients with negative plain radiographs because it has been shown to have high sensitivity (94 to 100%).14 ^– 16 However, when compared with other imaging modalities, it has a high false-positive rate (up to 25%) and a low specificity (varying between 60 and 95%) due to increased uptake from other traumatic conditions, such as scapholunate instability, bone bruises, synovitis, and arthritis.17 ^– 23 Bone scanning does not reach maximal predictive power until 72 hours following the injury, and subsequently in the earliest period, bone scintigraphy may be negative.24 ^, 25 A negative bone scan after 72 hours rules out a scaphoid fracture, which is why it has been popular. A positive scan in an otherwise healthy individual diagnoses an injury but does not specifically diagnose a fracture. A positive scan often requires further investigation, such as a longitudinal CT scan, to confirm a fracture.26 In 2005, Groves et al performed quantification bone scans on patients with suspected scaphoid fractures to determine if it was possible to accurately state whether a “hot spot” seen on scintigraphy was likely to be the result of a scaphoid fracture versus other pathology.7 Quantification bone scanning involves assigning numerical values to the degree of radioisotope (^99mTc-MDP) uptake in hot spots ( Fig. 4.3 ). Overall greater values were found in hot spots that were shown to be the result of a scaphoid fracture, based on CT images, versus other pathology.

Computed Tomography

The high resolution of modern CT scanners, particularly for osseous structures, makes CT ideal to detect occult fractures, to determine the direction of displacement of carpal fractures, and to evaluate fracture healing.27 Temple et al have demonstrated that CT is superior to plain radiographs in the identification of fracture displacement.28 It is well recognized that although the resolution of CT is lower than that of MRI, its spatial resolution is superior, and therefore CT is better able to detect displacement.29 CT as an imaging modality is therefore frequently used in attempting to diagnose a scaphoid fracture that is radio-occult on plain radiographs, and also in further preoperative assessment of a previously diagnosed fracture.

CT in the longitudinal axis of the scaphoid is now the preferred technique because it provides a greater appreciation of the anatomy and deformity, including any humpback deformity.31 ^, 32 The patient lies prone on the CT table with the affected arm above the head and the wrist in radial deviation and neutral flexion. The scanning plane is then oriented along the axis of the first metacarpal ( Fig. 4.4A ). If the correct orientation is obtained, equal portions of the proximal and distal poles of the scaphoid will be visualized on either side of the capitate. This is the “target sign” that appropriate orientation has been obtained30 ( Fig. 4.4B ).

The American College of Radiology has devised a scaling system where by various imaging modalities are given a score from 1 to 9 (with 9 the highest) for their effective-ness in detecting specific pathology.9 CT of the wrist in suspected acute scaphoid fractures with initially normal plain films has been given an appropriateness criterion score of 4 out of 9.9 This lower score is due to the fact that other associated injuries, such as scapholunate ligament tears, can be missed with CT. For these reasons, MRI is the preferred imaging modality in patients with a suspected fracture and normal radiographs. However, once a scaphoid fracture is confirmed, longitudinal CT is the best modality to assess the details of the fracture. This includes fracture position, displacement, angulation, and union status.3 ^, 31 ^, 32 CT provides the best appreciation of the morphology of the fracture ( Fig. 4.5 ).1 ^, 28 ^, 33 Because of the excellent resolution provided by longitudinal CT, fine details such as comminution, sclerosis, and orientation are appreciated much more clearly than with any other imaging. CT provides a level of detail such that the surgeon can understand the “personality of the fracture.”33