Search in the Abstract Database

Search Abstracts 2016

* = Presenting author

P122 Role of diffusion-weighted imaging in MRI-enterography for the evaluation of surgical risk in patients with Crohn’s disease

A. Testa*1, A. Rispo1, P. Mainenti2, D. Musto1, N. Imperatore1, M. Rea1, O.M. Nardone1, M. L. Taranto1, F. Castiglione1

1University ‘Federico II’ of Naples, Department of Clinical Medicine and Surgery, Naples, Italy, 2University ‘Federico II’ of Naples, Naples, Italy


In Crohn’s disease (CD), it is useful to discriminate inflammatory from fibrotic lesions. MRI-diffusion-weighted imaging (DWI) is able to identify active inflammation in most pathological tissues. Aim: To define the role of DWI in evaluating the risk of surgery in CD.


We performed an observational prospective study including all consecutive CD patients with active CD undergone MRI. MRI study included measurement of bowel wall thickness (BWT), CD extension, enhancement pattern, pre-stenotic dilation, presence of oedema and/or comb-sign, presence of fistulas/abscesses, and T2 and T2 sequences. Further, all patients were studied using DWI sequences defining visual analysis of iperintensity and analysis of apparent diffusion coefficient (ADC) maps. Statistical analysis was performed dividing all patients in 2 groups (operated vs not operated) using T-student and X-square test when indicated. To identify the variables associated to surgical risk, we performed a logistic multiple regression expressing the risk in terms of odd ratio. Finally, the diagnostic accuracy was tested using a ROC curve.


Enrolled were 110 patients; 127 bowel segments resulted pathologic at MRI. During the follow-up period, 26 patients (23.6%) and 31 segments were resected. At all pathological segments, the iperintensity in DWI sequences, the reduction of ADC max, ADC medium, and the presence of fistulas/abscesses were significantly associated with a subsequent surgical intervention (p < 0.05). In particular, the presence of CD complication was the variable with the highest risk of surgery (p = 0.008; OR 3.9; 95% CI 1.4–10.7). When excluding the patients with complications, we reported a significant correlation of DWI iperintensity, ADC max and medium with surgical intervention. The reduction of ADC medium was the variable with the highest risk of surgery (p = 0.03; OR 2.0; 95% CI 0.79–0.92). The cut-off value for discriminating patients at risk of surgery was 1 081 x 10-3 mm2/s (sensibility 55.6%, specificity 70.3%, PPV 33.3%, and NPV 85%).


The presence of fistulas/abscesses remains the variable most associated with surgical risk in CD. In not complicated CD, DWI sequences at MRI-Enterography correlates with the need of surgery. High value of ADC medium shows high NPV for surgery in CD patients.