Endgames Picture Quiz

A young man with wrist pain

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e4466 (Published 13 July 2012) Cite this as: BMJ 2012;345:e4466
  1. Hamoun Rozati, foundation year 1 trainee, trauma and orthopaedics,
  2. Zahra N Jaffer, foundation year 1 trainee, trauma and orthopaedics,
  3. Nawfal Al-Hadithy, core trainee year 2, trauma and orthopaedics
  1. 1Lister Hospital, Stevenage, SG1 4AB, UK
  1. Correspondence to: H Rozati h.rozati{at}nhs.net

A 24 year old left handed builder presented to the emergency department after falling on to an outstretched hand two days ago while playing football. Since then he had been experiencing increasing pain in his wrist.

He was otherwise fit and well with no medical problems.

On examination his wrist was tender in the anatomical snuff box. He was neurovascularly intact distally and had a full range of movement. There was increased pain on dorsiflexion but no obvious swelling, bruising, or deformity. He had no other injuries.

A radiograph was taken of his right wrist (fig 1).



  • 1 Describe the radiographic image. What is the diagnosis?

  • 2 What clinical tests support this diagnosis?

  • 3 How would you initially manage this patient?

  • 4 How would you manage this patient in the longer term?

  • 5 What complications might be expected?


1 Describe the radiographic image. What is the diagnosis?

Short answer

This is an anterioposterior radiograph of the right wrist showing a fracture of the waist of the scaphoid, which appears undisplaced.

Long answer

This is an anterioposterior radiograph of the right wrist showing a fracture of the waist of the scaphoid, which appears undisplaced.

The scaphoid bone lies radially in the proximal carpal row and is the most commonly fractured carpal bone.1

Three fracture sites of the scaphoid are commonly recognised (fig 2):

  • Fractures through the waist (the narrowest part of the scaphoid and the most commonly injured part)

  • Fractures through the proximal pole

  • Fractures through the tubercle.1


Fig 2 Anterioposterior radiograph of the right wrist showing the three sites of fracture of the scaphoid: (A) a fracture of the tubercle, (B) a fracture through the waist of the scaphoid, and (C) a fracture of the proximal pole

The rates at which these sites are fractured are 70%, 20%, and 10%, respectively.2

Although the fracture does not seem to be displaced, plain radiographs are poor at detecting small degrees of displacement (1 mm), with low sensitivity and significant interobserver variability.3 Computed tomography is more sensitive at determining whether displacement is present.4 As a result, some clinicians request computed tomography for any scaphoid fracture seen on plain radiography to determine whether it is displaced.

2 What clinical tests support this diagnosis?

Short answer

A scaphoid fracture is suggested by tenderness in the anatomical snuff box, tenderness over the scaphoid tubercle, a positive scaphoid compression test (longitudinally compress the patient’s thumb along the line of the first metacarpal to reproduce pain), and pain in the anatomical snuff box on pronation of the forearm.5

Long answer

A scaphoid fracture is commonly suspected when wrist pain follows a fall on to an outstretched hand (commonly abbreviated as FOOSH). The clinical signs, with their respective sensitivities, specificities, positive predictive values, and negative predictive values, have been documented (table).6

Sensitivities, specificities, positive predictive values (PPVs), and negative predictive values (NPVs) of clinical signs in detecting a scaphoid fracture6

View this table:

Despite many of the tests having high sensitivities and negative predictive values, their specificities tend to be low,5 so although they are useful in detecting suspected scaphoid fractures, in isolation their ability to rule out the diagnosis is poor.

A combination of these clinical signs may therefore be the best method of increasing the specificity of the clinical examination for a suspected scaphoid fracture. One study found that, when used in combination, testing positive for anatomical snuff box tenderness, having tenderness over the scaphoid tubercle, and eliciting a positive scaphoid compression test within the first 24 hours of the injury had 100% sensitivity and 74% specificity for diagnosing a scaphoid fracture.7

3 How would you initially manage this patient?

Short answer

Provide analgesia, immobilise the wrist, and refer to the fracture clinic with a note that he should be seen within seven days.

Long answer

The wrist should be immobilised for pain relief and to prevent further displacement of the fracture and non-union. The patient will need to be followed up in a fracture clinic within seven days.

The ideal method of immobilisation is unclear. A recent meta-analysis that compared below elbow and above elbow casts, incorporation and non-incorporation of the thumb into the cast, and whether the wrist should be held in any degree of flexion or extension failed to show that any one option was better than another in terms of union rate, pain resolution, grip strength, or time to union.8

Often there is a high index of suspicion for a scaphoid fracture on clinical examination but the initial plain radiograph shows no abnormalities. If a scaphoid fracture is suspected, the wrist should be immobilised and the patient referred to a fracture clinic to be seen within one week for further evaluation. If at follow-up the patient is still symptomatic then magnetic resonance imaging is recommended because repeat radiographs have been shown to be unreliable.9 Magnetic resonance imaging can obtain images in any plane, which are sensitive to oedematous changes and are free from the streak artefacts that may obscure plain radiographs.

4 How would you manage this patient in the longer term?

Short answer

Undisplaced fractures are typically managed conservatively and immobilised for a minimum of six weeks, although some surgeons advocate early surgical intervention. In either case follow-up is needed to assess for complications.

Long answer

There is considerable debate as to whether undisplaced or minimally displaced scaphoid fractures should be treated conservatively or operatively.

Conservative management comprises immobilisation in a scaphoid or Colles’ plaster for six weeks, which is sufficient to achieve bony union in more than 80% of cases.10 Operative management typically comprises the use of headless compression screws, such as the commonly used Herbert screws. Vascular bone grafts may also be used at the time of surgery and can help prevent the development of avascular necrosis, although such grafts are usually reserved for confirmed cases of non-union.11

Although operative intervention increases the risk of infection, advocates cite advantages of early operative fixation, lower rates of non-union, an earlier return to work, and less chance of osteoarthritis. However, two systematic reviews found no significant difference in union rate, return to work, grip strength, range of motion, or patient satisfaction between operative and non-operative treatment for undisplaced fractures and concluded that, given the increased chance of infection, surgery could not currently be recommended.12 13

Surgery is often recommended for displaced unstable fractures because of the increased risk of malunion or non-union. Unstable fractures are those with a displacement of more than 1 mm or an angulation of more than 15° between the fragments. Additional fractures, trans-scaphoid perilunate dislocations, and multi-fragment fractures are also classified as unstable.14

A union rate of only 82% has been reported for treatment of displaced fractures in a below elbow cast alone,15 whereas a union rate of 93% has been reported for fixation with Herbert screws.16 However, a meta-analysis of current evidence supporting operative fixation versus casting for acute scaphoid fractures found that the current literature is underpowered and cannot provide any definitive conclusions.13

5 What complications might be expected?

Short answer

The main problem is non-union, which can be best assessed by computed tomography. Avascular necrosis is also a possibility.

Long answer

The high incidence of non-union in scaphoid fractures arises from the fact that the scaphoid cannot undergo secondary healing. It is covered almost entirely by articular cartilage and so lacks the periosteum needed to form a callus.17 The blood supply to the scaphoid is also important in understanding potential complications after a fracture. The scaphoid is supplied by the scaphoid branches of the radial artery, which enter the distal aspect of the bone before extending proximally. This blood supply is easily disrupted by a fracture, and this adversely affects union rates and the likelihood of avascular necrosis. In addition, because the blood supply enters distally before extending proximally, the more proximal the fracture site the greater the likelihood that the proximal pole of the scaphoid will undergo avascular necrosis.18

Patients with confirmed scaphoid fractures must have regular follow-up to monitor for complications. Failure to prevent non-union may lead to a permanent flexion of the distal pole of the scaphoid and a so called humpback deformity. With this deformity the ability to perform wrist extension is generally lost because of associated carpal collapse.1

Serial radiographs are typically used to assess healing even though there is significant interobserver variation in the assessment of these fractures.19 Computed tomography, however, is a more accurate method of diagnosing complications, with higher interobserver and intraobserver reliability (fig 3).20


Fig 3 Computed tomogram showing non-union of a scaphoid fracture (dotted arrow)

Magnetic resonance imaging can sometimes be useful in assessing the healing of scaphoid fractures (fig 4). Although it can confirm bony union in most patients, occasionally it shows abnormal signal intensity around a stable fracture even as healing progresses to union. The only definitive sign of union is the return of normal marrow continuity across the fracture line.21 This suggests that computed tomography is a better imaging modality for assessing bone healing, whereas magnetic resonance imaging may be useful in suspected avascular necrosis.


Fig 4 Magnetic resonance imaging showing enhancement of the proximal and distal scaphoid but no enhancement at the site of the fracture—that is, no obvious evidence of healing granulation tissue (dotted arrow)

Patient outcome

Our patient had no evidence of union after 12 weeks of immobilisation in a Colles’ cast. The pain was beginning to affect his ability to perform everyday tasks, and he had been off work since the time of first presentation.

He underwent surgery for screw fixation with bone grafting; avascular grafts were used because magnetic resonance imaging showed no signs of avascular necrosis. His wrist was kept immobilised for a further six weeks after surgery, and a follow-up appointment in the fracture clinic was made to assess his progress.

When he returned to the fracture clinic the pain in his wrist was greatly reduced. The cast was removed and the range of motion in his wrist was much better. He was encouraged to keep mobilising his wrist and to return if he had any problems. He has now returned to full time work and started playing sports again.


Cite this as: BMJ 2012;345:e4466


  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Not commissioned; externally peer reviewed.

  • Patient consent obtained.


View Abstract