• 
• 

Three Part Question

In [adult patients with a suspected scaphoid fracture], is [MRI is better than bone scintigraphy (BS)] in [diagnosing a fracture with an initially negative X-ray result]?

Clinical Scenario

A 25-year-old right-handed carpenter came to the Emergency Department (ED) with a recent history of a fall onto his outstretched hand. He complained of pain in the wrist and some tenderness was reported in the anatomical snuffbox. The subsequent X-ray showed no abnormality. Should the patient undergo another investigation, such as an MRI, there and then?

Search Strategy

Ovid MEDLINE(R) 1950 to June Week 4 2010
[exp Scaphoid Bone OR exp Carpal Bones OR scaphoid$.mp. OR exp Fractures, Bone OR fracture$.mp]
Limit to English language AND humans AND all adult (19 plus years).
AND
[exp Magnetic Resonance Imaging OR magnetic resonance imaging$.mp. OR MRI$.mp. OR MR scan$.mp.]
AND
[exp Radionuclide Imaging OR bone scintigraphy$.mp. OR bone scan$.mp.]

Search Outcome

18 papers were identified in Medline, five papers were relevant to the three-part question.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Tiei-van Buul MMC et al. 1996 The Netherlands	16 patients (11 male, 5 female), mean age: 36 years old Presented to ED with suspected scaphoid fracture and negative initial x-ray	Single blinded, Diagnostic cohort study	Scaphoid fracture	5 MRI vs 5 BS	Small sample size. No statistical analysis included. Publication bias not mentioned. Single blinded, may be bias. No sensitivity or specificity included.
			False positive	2 BS
			False negative	1 MRI
			MRI agrees with BS	11 of 16
Thorpe AP et al. 1996. Scotland.	59 patients (36% male, 64% female), mean age: 22 years old Presented to ED with suspected scaphoid fracture, negative initial x-ray and 10th day review radiograph.	Prospective, Double blinded diagnostic cohort study	Scaphoid fracture	4 MRI vs 4 BS	Publication bias not mentioned. Statistical analysis was not included. Relatively small sample size. Result comparison with previous studies was not included. Sensitivity or specificity not included in numbers.
			False positive	1MRI vs 3 BS
			Cost effectiveness	£67 MRI vs £104 BS
Kitsis C et al 1998. England	22 patients (9 males, 13 females), mean age: 34 years old Patients referred from ED and other orthopaedic clinics to a special hand clinic with suspected scaphoid fracture and negative initial and follow-up x-ray at 2 weeks post injury	Prospective study, Diagnostic cohort study	Scaphoid fractures	3 MRI vs 3 BS	Small sample size. Numbers do not add up. Publication bias not mentioned. Statistical analysis was not included. One patient had normal BS but missed MRI appointment. No sensitivity or specificity included.
			False positive	1 BS
			MRI agrees with BS	17 of 22
			Cost effectiveness	£200 MRI vs £80 BS
			Soft tissue injuries	4 MRI vs 0 BS
Fowler C et al. 1998 Wales.	43 patients (21 males, 22 females), average age: 32 years old Presented to ED with suspected scaphoid fracture, negative initial x-ray, having MRI and BS at 7-10 days post-injury	Prospective diagnostic cohort study	Scaphoid fracture	6 MRI vs 6 BS	Publication bias not mentioned. Statistical significance of results was not included. Relatively small sample size.
			False positive	2 BS
			False negative	1 BS
			MRI agrees with BS	40 of 43
			Specificity	100% MRI (CI: 91-100) vs 95 % BS (CI: 89-99.9)
			Sensitivity	100% MRI (CI: 54-100) vs 83% BS (CI: 86-99.9)
Beeres FJP et al. 2008. The Netherlands	100 patients (50 male, 50 female), mean age: 42 Presented to ED with suspected fractured scaphoid but negative initial radiograph finding	Prospective study, Blinded Diagnostic cohort study	Early MRI (within 24 hrs of injury)	Sensitivity: 80% Spesificity: 100%
			BS 3-5 days post-injury	Sensitivity: 100% Specificity: 90%

Comment(s)

The results are contradictory on the subject of which method is superior, with three out of the five studies claiming that MRI is better for diagnosing occult scaphoid fracture. However, it can be concluded that MRI is better in terms of detecting soft tissue injuries within 14 days of trauma, while BS is more specific and sensitive to bone fractures, even at an early stage post-injury. A similar BET has been developed with the conclusion that MRI is superior to BS.21 However, the conclusion has been challenged by a recent paper included here showing stronger evidence that BS is better than MRI at diagnosing scaphoid fractures.

Clinical Bottom Line

BS has shown better sensitivity and specificity than MRI at diagnosing occult scaphoid fractures with a normal initial radiograph. BS may also be used in patients with claustrophobia. BS and MRI may be used in conjunction to diagnose wrist injury. As cost-effectiveness and safety measures for radiation of BS have not adequately been established, further studies must be done to answer these questions.

References

1. Tiei-van Buul MMC, Roolker W, Verbeeten Jr BWB, Broekhuizen AH Magnetic resonance imaging versus bone scintigraphy in suspected scaphoid fracture. Eur J Nucl Med 1996; 23: 971-5.
2. Thorpe AP, Murray AD, Smith FW, Ferguson J Clinically suspected scaphoid fracture: a comparison of magnetic resonance imaging and bone scintigraphy The British Journal of Radiology 1996; 69: 109-13.
3. Kitsis C, Taylor M, Chandey J, Smith R, Latham J, Turner S and Wade P Imaging the problem scaphoid Injury 1998; 29 (7): 515-20
4. 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 Radiol 1998; 27:683-7
5. Beeres FJP, Rhemrev SJ, den Hollander P, Kingma LM, Meylaerts SAG, le Cessie S, Bartlema KA, Hamming JF, Hogervorst M Early magnetic resonance imaging compared with bone scintigraphy in suspected scaphoid fractures J Bone Joint Surg [Br]. 2008; 90-B:1205-9.