Published 12/1/2010

MRI can yield false positives for scaphoid

Magnetic resonance imaging (MRI) is often advocated as a way to evaluate patients with suspected scaphoid fractures. But according to the results of a study presented by Andele D. de Zwart, MS, a medical student at Medical Centre Haaglanden, The Hague, Netherlands, MRI should not be used as a reference standard because it is not 100 percent specific for diagnosing this injury and can yield false positives in healthy volunteers.

In their prospective study, Mr. de Zwart and his fellow researchers recruited 32 healthy volunteers and scanned both of their wrists. Researchers mixed the 64 scans of wrists lacking scaphoid fractures with 60 scans of patients who had normal scaphoid radiographs but were clinically suspected of having scaphoid fractures. The 124 MRI scans, which were blinded and randomly ordered, were independently reviewed by five radiologists using a standard protocol to determine whether a scaphoid fracture was present.

On average, the five radiologists diagnosed 2.6 fractures in healthy volunteers, indicating a specificity (proportion of healthy volunteers correctly diagnosed as having no fracture) of 96 percent (95 percent CI: 0.94–0.98). Moderate interobserver agreement was indicated statistically by the multi-rater kappa of 0.44 (p < 0.001).

The researchers noted that “the use of a reliable reference standard for no fractures in healthy volunteers allows us to estimate the specificity of MRI for scaphoid fracture as approximately 96 percent, which is good, but not perfect. This is important because patients with suspected scaphoid fractures have a low prevalence of true fractures.” The investigators added that “even small imperfections in diagnostic tests are magnified” in these patients.

Coauthors for “False Positive MRIs for Scaphoid Fracture in Healthy Volunteers” include Frank J. P. Beeres, MD, PhD; Sven Meylaerts, MD, PhD; Daan Otis, Jr.; David C. Ring, MD, PhD; Steven Rhemrev, MD.

Disclosure information: Mr. deZwart; Drs. Beeres, Meylaerts, Otis, and Rhemrey—no conflicts; Dr. Ring—Wright Medical Technology, Inc., Acumed LLC, Biomet, Illuminos, Mimedex, Stryker.

Bottom line

  • This prospective study featured an independent review of MRIs of healthly volunteers and patients with scaphoid fractures to determine the usefulness of MRI as a diagnostic tool.
  • Radiologists diagnosed 2.6 fractures on average in healthy volunteers.
  • This translates to a 96 percent specificity in the use of MRI to identify scaphoid fractures.