increased opacity (sclerosis) and a frank fracture line may be identifiable at a later time point [21–24]. The reactive bone is generally confined to a small area and usually involves only one cortical surface. Ultimately, the area of periosteal reaction thickens and the fracture line, if seen previously, disappears [21]. Following trabecular microfracture, osteoblasts lay down new bone along the injured trabeculae. Depending upon the timeline, this may produce subtle blurring of the trabeculae with faint increased radiopacity and later thickened trabeculae producing more evident sclerosis secondary to peri‐trabecular callus [20, 25]. Trabecular bone is reported to have a metabolic turnover eight times faster than cortical bone [26], leading to the possibility that subtle changes may be identified in this location first.
In man, the limitations of conventional radiographs for detection of stress injuries are well documented [2227–29]. The multifaceted variables in the continuum of the stress response account for the variation in radiographic appearance [30]. Given the microscopic remodelling that occurs in the early stages of a stress injury, the overall sensitivity of radiographs can be low and findings may be reserved until the healing phase, 12–21 days [14, 20,31–34] and in some instances four to six weeks [35] after a stress fracture, has occurred (see Figure 5.10a and c). Advances in radiography since the digital era have made subtle changes easier to recognize, but sometimes fractures never become radiographically apparent [34,36–39].
If a fracture is identified, this provides a risk bracket for the patient which assists with management strategies and whether the nature of the changes supports radiographic monitoring. If the radiographs are negative, depending upon the area, it may be advisable to repeat radiographic examination in 7–14 days and/or consider an alternative imaging modality.
Articular Fractures
A fracture is considered articular if it communicates with a joint. From a radiographic perspective, this involves discontinuity in subchondral bone and by implication overlying cartilage. A high index of suspicion for articular involvement can be raised with the presence of synovial distension. Articular involvement can have a major impact on case management and prognosis and, when suspected, radiographs should be carefully scrutinized using lesion‐oriented oblique projections.
Slab fractures connect two, usually proximal and distal, articular surfaces of cuboidal bones. Third carpal and central and third tarsal bones are most commonly affected. Dorsoproximal–dorsodistal (skyline) radiographs of the proximal and distal rows of the tarsal bones are not possible, which can make identification and determination of configuration difficult particularly with respect to the central tarsal bone. CT has made a major contribution to the database of injuries, and knowledge of common configurations aids radiographic evaluation (Chapter 29).
Fissure Fractures
Fissure fractures are unicortical or involve a single subchondral bone plate. Beam angle is critical to identification, and multiple slight variations in projection orientation should be made if a fracture is suspected but not identified (Figure 5.4a and b).
Avulsion Fractures
Avulsion fractures represent disruption of all or part of an enthesis. They can happen at any location and may be monotonic or fatigue related. The radiographic findings are related to the area of involvement and time frame of the injury.
Compression Fractures
Acute, minimally displaced, compression fractures can be difficult to identify, and time for associated osseous resorption and/or callus production may be necessary for confident diagnosis.
Accompanying Features
Soft Tissue Swelling
The degree and nature of soft tissue swelling, whether intra‐capsular, extra‐capsular, focal or diffuse, with accompanying effacement of facial planes or fat pads, can help focus on a region of interest. For example, radiographs of a transverse stress fracture of the distal third metacarpal bone may initially reveal no osseous disruption and show only a subtle adjacent soft tissue swelling. Over ensuing weeks, the radiographs can progress dramatically (Chapter 22). Following a skull fracture, haemorrhage in the guttural pouches can obliterate the normal gas lucency which is replaced by soft tissue opacity or produce a fluid line secondary to a gas–fluid interface. Accompanying features of ventral deviation of the dorsal pharyngeal wall and dorsoventral attenuation of the nasopharynx may also be visible.
Presence of Gas Lucency
Open fractures are most commonly secondary to impact injuries. The presence of an open wound can be seen radiographically as disruption of and/or defects in the normal soft tissue opacity with varying degrees of gas opacity inclusions. Gas opacity extending to the fracture on two orthogonal projections suggests direct communication with the wound. A gas cap in the most proximal extent of a synovial cavity raises the suspicion of synovial penetration (see Figure 26.18). In the absence of a wound, gas in a joint can be explained by vacuum phenomena [40]. The authors have seen this occasionally following third metacarpal bone condylar fractures that have had concomitant joint capsule tearing identified during arthroscopy. Gas can also accumulate in the subarachnoid and cervical epidural spaces following some basilar skull fractures [41]. Frontal or sphenopalatine sinus fractures or a fractured petrous temporal bone in combination with a ruptured tympanic membrane can also lead to free gas within the calvarium [10, 41].
Monitoring Fracture Healing
One of the most obvious but salient requirements of follow‐up radiographs is that the images must be comparable to those taken previously. Small changes in position can result in the X‐ray photon beam not being parallel to the fracture plane. Endosteal and periosteal new bone formation can also appear to be reducing or increasing. Both errors lead to incorrect conclusions which in turn can compromise case management.
Healing and remodelling of fracture margins occur simultaneously. Even with internal fixation and primary healing (Chapter 6), there is often initial resorption along the fracture line. It is important to establish an expected time frame for uncomplicated fracture healing for individual sites. For example, bone in and adjacent to the proximal subchondral bone plate can take the longest to heal in parasagittal proximal phalangeal fractures [42]. Awareness of common accompanying features is also needed. For example, periosteal new bone formation on the dorsal proximal aspect of the proximal phalanx frequently extends further distad than radiographically identifiable fracture lines [43].
Surgical implants are examined carefully for evidence of migration, bending, breakage or adjacent osseous lucency, which may suggest instability or infection (Figure 14.5c). Care must be taken to differentiate abnormality from Uberschwinger artefact. With healing, adjacent soft tissues should exhibit reduced swelling and more clearly defined fascial planes. Persistent swelling whether generalized or focally over an implant generally warrants further scrutiny (see Figure 14.5a).
In articular fractures, the cartilage space, articular margins, subchondral bone and entheses are evaluated for evidence of reactive or degenerative changes. Resolution or persistence of intra‐articular fat pad effacement in applicable joints provides a guide to joint distension.
In the first week or two of second intention healing (Chapter 6), there is an initial loss of mineral density adjacent to the fracture resulting in reduced sharpness of the margins and a possible increase in the fracture gap. It usually takes 10–12 days for endosteal and periosteal new bone formation to become evident. Within 30 days, the fracture line should be less distinct and callus demonstrate