Staging of hip avascular necrosis: is there a need for DWI?

Background: No comprehensive study has been performed to stage avascular necrosis of the hip using diffusion-weighted imaging (DWI). Purpose: To determine apparent diffusion co-efficient (ADC) alterations in hip avascular necrosis (AVN) and to determine variations of ADC values according to stages of disease. Material and Methods: The study is approved by our institutional review board and local ethical committee. Written informed consent was present for each subject. Thirty-five femoral heads of 21 cases affected by AVN were included in the study. Control group consisted of both femoral heads of 10 healthy volunteers. The hips affected by AVN were staged according to Ficat and Arlet classification system from I to IV. All cases underwent to routine hip magnetic resonance imaging (MRI) and DWI performed with a single-shot fast spin echo sequence at a b value of 600 s/mm 2 . The ADC values were calculated automatically by placing ROIs on AVN lesions in affected patients and both femoral heads of control group. The median ADC value obtained from femoral heads of control group and that from AVN lesions were compared by Mann-Whitney U test. The median ADC values of AVN lesions at different stages were compared by Kruskal-Wallis test. Results: The median ADC value of normal bone measured in control group was 185.5 ± 133.2 x 10 -6 mm 2 /s. The median ADC value measured in hip avascular necrosis lesions was 988.0 ± 332.7 x 10 -6 mm 2 /s. ADC values in hip AVN lesions were statistically significantly higher than normal bone marrow (P 0.05). Conclusion: Although DWI is a promising imaging tool that provides valuable diagnostic information in hip AVN, it fails to distinguish between different stages, and therefore is of limited value.