Diagnosis of osteomyelitis by imaging can be challenging. The
feasibility of diffusion-weighted imaging (DWI) as ancillary
sequence was evaluated in this study.
To evaluate DWI for differentiation between osteomyelitis, bone
marrow edema, and healthy bone on forefoot magnetic resonance
A total of 60 consecutive patients undergoing forefoot MRI divided
into three study groups (20 subjects each): osteomyelitis, bone
marrow edema, and healthy bone.
A 1.5T and 3T MRI scanners; readout-segmented multishot echo planar
Two independent radiologists measured apparent diffusion coefficient
(ADC) values within abnormal or healthy bone.
ADC values were compared between groups (pairwise t-test with
Bonferroni-Holm correction for multiple testing). Intraclass
correlation coefficient (ICC) was calculated to assess inter-reader
agreement. Threshold ADC values were determined as the cutoffs that
maximized the sum of sensitivity and specificity. Receiver operating
characteristic (ROC) analysis was performed with statistical
threshold of P < 0.05.
Inter-reader agreement was 0.92 in the healthy bone group and 0.78
in both the edema and osteomyelitis groups. Average ADC values were
significantly different between groups:
1432 ± 222 × 10 mm /sec
1071 ± 196 × 10 mm /sec
(bone marrow edema), and
277 ± 89 × 10 mm /sec
(healthy bone). A threshold ADC value of 534 × 10
mm /sec distinguishes between healthy and abnormal bone with
specificity and sensitivity of 100% each. For distinction between
osteomyelitis and bone marrow edema, two cutoff values were
determined: a 95%-specificity cutoff indicating osteomyelitis
(>1320 × 10 mm /sec) and a
95%-sensitivity cutoff indicating bone marrow edema
(<1155 × 10 mm /sec). Diagnostic accuracy
of 95% was achieved for 73% (29/40) of the subjects.
DWI with ADC maps distinguishes between healthy and abnormal bone on
forefoot MRI. Calculated cutoff values allow confirmation or
exclusion of osteomyelitis in a high proportion of subjects.
2 TECHNICAL EFFICACY: Stage 2.