Purpose: To assess the reliability of MRI and scintigraphy with 99mTc-HMPAO-labelled granulocytes in the diagnosis of bone infection.
Materials and methods: Between November 2001 and November 2003, 25 patients (16 men and 9 women; age range 22-72 years; mean age 48 years) with suspected bone infection were evaluated. The lower limbs were more frequently involved (20/25 cases). MRI (T1 and T2-w sequences, both with and without fat suppression; T1-w fat-suppressed sequence after Gadolinium administration) and scintigraphy with 99mTc-HMPAO-labelled granulocytes were performed in all patients; the study was performed 30 min, 150 min and 24 h after the injection of 99mTc labelled autologous granulocytes. The maximum interval between MRI and scintigraphy was 15 days. The diagnosis was confirmed surgically (18/25 cases) or by clinical follow-up (7/25 cases).
Results: In 10/25 cases the clinical suspicion of bone infection was confirmed by MRI, scintigraphy and surgery. In 12/25 cases the clinical suspicion of bone infection was not confirmed either by imaging, surgery (5/12 cases) or follow-up (7/12 cases). In 3/25 cases the results were discordant: in 1/3 cases with chronic osteomyelitis confirmed by MRI and surgery, a false-negative diagnosis was made by scintigraphy. In 2/3 cases with negative MRI diagnoses confirmed by surgery, a false-positive diagnosis was made by scintigraphy owing to the infection of adjacent soft tissues.
Conclusions: MRI is more reliable for the detection and evaluation of the local spread of bone infection. Conversely, when metallic devices causing artefacts on MR images are present, scintigraphy is the preferred alternative modality.