• 
• 

Three Part Question

In an adult with a [clinically suspected scaphoid fracture] is [magnetic resonance imaging better than bone scintigraphy] at [reaching a diagnosis]?

Clinical Scenario

An adult presents to the review clinic two weeks after falling onto his outstretched hand. A scaphoid fracture was suspected but no fracture was seen on plain scaphoid views. He continues to have scaphoid tenderness. You wonder whether a magnetic resonance scan would be better than a bone scan in confirming or excluding a scaphoid fracture.

Search Strategy

Medline from 1966 until March 2005
Embase 1974 – March 05
Cochrane Library 2005, Issue 1.
Medline - (fracture$.af. OR exp Fractures, Comminuted/ OR exp Fractures, Closed/ OR exp Fractures/ OR exp Fractures, Ununited/ OR exp Fractures, Malunited/) AND (exp Scaphoid Bone/ OR scaphoid.af. OR exp Carpal Bones/) AND (exp Magnetic Resonance Imaging/ OR (magnet$ ADJ reson$ ADJ imag$).mp. OR mri.af. OR exp Magnetic Resonance Spectroscopy/ OR (magnet$ ADJ reson$ ADJ spect$).mp. OR nmr.af.) AND (exp Radionuclide Imaging/ OR exp Technetium/ OR exp Technetium Tc 99m Medronate/ OR scintigraph$.af. OR bone scan.af.)
Embase - ((fracture$.af. OR fracture#.w..de. OR Hand-fracture#.de. OR hand-injury#.de. OR fracture-nonunion#.de.) OR (scaphoid-fracture#.de.)) AND (Nuclear-Magnetic-Resonance-Imaging#.de. OR (magnet$ ADJ reson$ ADJ imag$).mp. OR (magnet$ ADJ reson$ ADJ spect$).mp. OR mri.af. OR nmr.af.) AND ((bone ADJ scan).mp. OR Bone-scintiscanning#.de. OR Medronate-Technetium-TC-99m#.de. OR Technetium-99m.#.de OR Whole-body-scintiscanning#.de. OR scintigraph$.af.))
Cochrane - Scaphoid [all fields] AND fracture [all fields]

Search Outcome

Altogether 23 papers were identified in medline, three of which were relevant to the three-part question, one further paper was found by searching the references.
30 papers were identified in Embase none relevant.
6 papers in Cochrane, none relevant.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Thorpe 1996 UK	61 patients (59 examinations) with signs of a scaphoid fracture but negative initial and 10 day follow-up X rays. Comparison of magnetic resonance and bone scans at 3-4 weeks after injury	Diagnostic cohort	Scaphoid fracture	4 detected by both MR and BS	Only 52% attended follow-up at 6 weeks for clinical evaluation and further X rays
			False positives	3 BS and 1 MR
			Other significant injuries	3 soft tissue injuries detected by MR
Tiel-van Buul 1996 Netherlands	19 patients with clinically suspected scaphoid fracture and negative initial X rays. Comparison of magnetic resonance and bone scan	Diagnostic cohort	Scaphoid fracture	5 on MR and BS	3 patients did not complete MR scan because of claustrophobia. Bone scan was the reference method.
			False positive	2 BS
			False negative	1 MR (retrospectively X ray positive)
			Agreement between BS and MR	11/16
Fowler 1998 UK	45 patients with clinically suspected scaphoid fractures with negative initial and 7-10 day follow-up X rays. Comparison of magnetic resonance and bone scan at 10-35 days after injury. Telephone follow-up at 12 months and then clinical evaluation and further X rays if indicated.	Diagnostic cohort	Scaphoid fracture	6 on MR and BS	2 patients did not complete MR scan because of claustrophobia. Only 8 patients followed up at 12 months (6 confirmed fractures and the 2 false positive on BS).
			False positive	2 BS
			False negative	1 BS
			Agreement between MR and BS	40/43
Kitsis 1998 UK	22 patients with suspected scaphoid fractures referred to a hand clinic after normal initial and 14 day follow-up X rays Comparison of magnetic resonance and bone scans at 2-4 weeks after injury. Follow-up at 8 weeks for clinical evaluation and further X rays	Diagnostic cohort	Scaphoid fracture	3 detected by both MR and BS	Small study. 1 patient had BS and then missed MR
			False positives	1 on BS
			Other significant injuries	4 significant soft tissue injuries detected by MR but not BS
			MR and BS in agreement	17/22

Comment(s)

Only 145 patients appear to have been entered into any sort of comparison between the two imaging modalities. Magnetic resonance imaging appears slightly superior in diagnosing occult scaphoid fractures. Magnetic resonance imaging also allows accurate diagnosis of clinically significant soft tissue injuries, which might otherwise be missed. It is also quicker to perform than a bone scan. However, some patients with claustrophobia will not tolerate it.

Clinical Bottom Line

Magnetic resonance imaging is the investigation of choice in the clinically suspected scaphoid fracture after negative initial and 10-14 day follow-up X rays. A bone scan is a reasonable alternative in patients with claustrophobia.

References

1. Thorpe AP, Murray AD, Smith FW, Ferguson J Clinically suspected scaphoid fracture: a comparison of magnetic resonance imaging and bone scintigraphy British Journal of Radiology 1996; 109-113
2. Tiel-van Buul MM, Roolker W, Verbeeten BW, Broekhuizen AH Magnetic resonance imaging versus bone scintigraphy in suspected scaphoid fracture European Journal of Nuclear Medicine 1996; 971-975
3. Fowler C, Sullivan B, Williams LA, McCarthy G, Savage R, Palmer A A comparison of bone scintigraphy and MRI in the early diagnosis of the occult scaphoid waist fracture. Skeletal Radiology 1998; 683-687
4. Kitsis C, Taylor M, Chandey J, Smith R, Latham J, Turner S, Wade P Imaging the problem scaphoid Injury 1998; 515-520