Abstract:
Introduction: The precise differentiation of pedal osteomyelitis versus other complications of patients with diabetes mellitus and occlusive arterial disease and the correct determination of the expansion represent a great diagnostic problem in all imaging methods.
Purpose: Evaluation of the value of dynamic MRI in comparison to conventional radiography, three-phase bone scintigraphy and static MRI in the diagnosis of pedal osteomyelitis
Methods: These 4 methods were applied prospectively in 40 feet of 38 patients. The diagnosis was confirmed through histology findings or clinical follow-up. Dynamic contrast-enhanced MRI was encountered with Flash-3D-sequences after intravenous injection of Gd-DTPA. With signal intensity-time curves the contrast enhancement rate was measured in different bone regions and soft tissue. To improve the interindividual comparability, a slope-factor was introduced to eliminate individual differences in pedal blood supply.
Results: 50% of the patients had pedal osteomyelitis. The overall accuracy of radiography, scintigraphy and static MRI was 70%, 80%, and 92,5%. Scintigraphy and MRI showed a sensitivity of 90%. Static MRI was specific in 95%. The accuracy of the combination of these three methods was 90%. Sensitivity, specificity and accuracy of dynamic MRI was 95%, 100%, and 97,5%. A slope value under 5%/min excluded an osseous infection. Values of more than 35%/min and a slope-factor over 18 confirmed an osteomyelitis.
Conclusions: Dynamic MRI is superior in diagnosis of pedal osteomyelitis in comparison to all other methods mentioned. It can exclude or confirm bone infection with extraordinarily precision. Because of the severe differentiation of vital, reactive, and infected bone the exact extension of the osteomyelitis can be determined.