Diagnostic performance of FDG-PET, MRI, and plain film radiography (PFR) for the diagnosis of osteomyelitis in the diabetic foot

Mol Imaging Biol. 2010 Jun;12(3):335-42. doi: 10.1007/s11307-009-0268-2. Epub 2009 Oct 9.


Background: The early and accurate diagnosis of osteomyelitis in the diabetic foot is essential to provide appropriate treatment and obviate long-term complications of the disease. The currently employed non-invasive imaging modalities such as plain film radiography (PFR) lack the sensitivity to accurately exclude osteomyelitis, while magnetic resonance imaging (MRI) is limited by its low specificity and contraindications in certain patients. Therefore, accurate non-invasive detection of osteomyelitis in the diabetic foot remains a challenge. [18F]-2-fluoro-2-deoxy-D-glucose (FDG)-positron emission tomography (PET) has been proven useful in other settings to detect infection. In this ongoing prospective study, we assessed the diagnostic performance of FDG-PET to diagnose osteomyelitis in the diabetic foot and compared it with that of MRI and PFR.

Methods: Patients who met the prespecified criteria for complicated diabetic foot underwent FDG-PET, MRI, and PFR of the feet. Each imaging study was then interpreted in a blinded fashion for presence of osteomyelitis or other abnormalities. The gold standard for diagnosis in each patient was based on surgical, microbiological, and clinical follow-up results.

Results: One hundred ten consecutive patients have been enrolled to date into this prospective project. FDG-PET correctly diagnosed osteomyelitis in 21 of 26 patients and correctly excluded it in 74 of 80, with sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of 81%, 93%, 78%, 94%, and 90%, respectively. MRI correctly diagnosed osteomyelitis in 20 of 22 and correctly excluded it in 56 of 72, with sensitivity, specificity, PPV, NPV, and accuracy of 91%, 78%, 56%, 97%, and 81%, respectively. PFR correctly diagnosed osteomyelitis in 15 of 24 and correctly excluded it in 65 of 75, with sensitivity, specificity, PPV, NPV, and accuracy of 63%, 87%, 60%, 88%, and 81%, respectively.

Conclusion: FDG-PET is a highly specific imaging modality for the diagnosis of osteomyelitis in diabetic foot and, therefore, should be considered to be a useful complimentary imaging modality with MRI. In the setting where MRI is contraindicated, the high sensitivity and specificity of FDG-PET justifies its use after a negative or inconclusive PFR to aid an accurate diagnosis.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Diabetic Foot / complications
  • Diabetic Foot / diagnostic imaging*
  • Female
  • Fluorodeoxyglucose F18*
  • Humans
  • Magnetic Resonance Imaging*
  • Male
  • Middle Aged
  • Osteomyelitis / complications
  • Osteomyelitis / diagnostic imaging*
  • Positron-Emission Tomography*
  • Radiography


  • Fluorodeoxyglucose F18