Evaluating the inflammatory activity in Crohn's disease using magnetic resonance diffusion kurtosis imaging

To explore the feasibility of diffusion kurtosis imaging (DKI) for evaluating inflammatory activity in Crohn's disease (CD).

In all, 51 CD patients were included, who were performed with consecutive enteroscopy, MR and DKI (b values = 0-2000 mm²/s). The lesions of bowel segments were graded as inactive (0-2), mild (3-6), and moderate-severe group (> 6) based on simplified endoscopic activity score for Crohn's disease (SES-CD), The abilities of the parameters of DKI and DWI in grading different activity lesions were compared.

One hundred and twenty-seven bowel segments including inactive (15), mild (45) and moderate-severe (67) were analyzed. ADC (r = - 0.627, p < 0.001), D_app (r = - 0.381, p < 0.001) and K_app (r = 0.641, p < 0.001) were correlated with SES-CD. These parameters were significantly different among the three groups (all p < 0.001). ROC analysis found ADC had the highest accuracy (AUC = 0.884, p < 0.001) to differentiate inactive from active group with the threshold at 0.865 × 10^-3 mm²/s, which was slightly higher than K_app (AUC = 0.867, p < 0.001) with the threshold at 0.645, and was obviously higher than D_app (AUC = 0.726, p = 0.005). Similarly, ADC also had the highest accuracy (AUC = 0.846, p < 0.001) to differentiate inactive-mild from moderate-severe group with the threshold at 0.825 × 10^-3 mm²/s, and minimally higher than K_app (AUC = 0.843, p < 0.001) with the threshold at 0.695, and obviously higher than D_app (AUC = 0.690, p < 0.001).

DKI is feasible and comparable to conventional DWI for the evaluation of inflammatory activity in CD.