1674
A new MR-based Perianal Crohn’s disease activity score: Multicentre study1Nottingham Digestive Diseases Centre, University of Nottingham, Nottingham, United Kingdom, 2Faculty of Applied Medical Sciences, Diagnostic Radiology, Jazan University, Jazan, Saudi Arabia, 3Sir Peter Mansfield Imaging Centre, University of Nottingham, Nottingham, United Kingdom, 4NIHR Nottingham Biomedical Research Centre at Nottingham University Hospitals NHS Trust and University of Nottingham, University of Nottingham, Nottingham, United Kingdom, 5Radiology, Nottingham University Hospitals NHS Trust, Nottingham, United Kingdom, 6Radiology, St Mark's Hospital and London North West Healthcare NHS Trust, London, United Kingdom
Synopsis
Perianal Crohn’s disease (pCD) is a potential complication in CD. Absence of reliable disease measures makes disease monitoring unreliable. MRI is an effective imaging method for the evaluation of patients with pCD. Quantitative MRI sequences, such as diffusion-weighted image (DWI), and magnetization transfer (MT) offer opportunities to improve diagnostic capability. The aim of this study was to measure disease activity within a pCD patient cohort using quantitative MRI sequences (DWI and MT), at different field strengths, before and after biological therapy. The study is ongoing with patients presenting with a range of clinical and inflammatory markers of disease activity.
Introduction
Perianal Crohn’s disease (pCD) is a potential complication of Crohn’s Disease [1]. Absence of reliable disease measures complicates monitoring of disease. Clinical measurements such as Perineal Disease Activity Index (PDAI) and Fistula Drainage Assessment are inaccurate with high inter-observer-variability[2]. These indices are therefore questionable when monitoring disease activity after medical therapy. MRI has been increasingly used for diagnosing and monitoring pCD patients, and has shown potential for assessing therapeutic response and inflammation. Quantitative MRI sequences, such as diffusion-weighted image (DWI), and magnetization transfer (MT) offer opportunities to improve diagnostic capability[3][4]. The aim of this study was to measure disease activity within pCD using quantitative MRI sequences (DWI and MT) at both 1.5T and 3T, before and after biological therapy.
Methods
This is an on-going prospective cohort study being conducted at Nottingham, London, Manchester, and Sheffield, aiming to recruit 25 patients with active pCD. Patients are scanned at 1.5T and 3T on two occasions, once before starting, and 12 weeks after starting biological therapy. To date, 23 patients have been recruited, however, three subjects were excluded and figure 1 provides the demographics. 9 patients have been fully analysed to date (Fig 1).
DWI, and MT sequences (parameters in Fig 2) were added to the standard clinical protocol of coronal, axial and sagittal T2-weighted turbo spin echo, oblique axial and oblique coronal fat-suppressed T2-weighted and pre/post contrast enhanced -T1 weighted scans (3T only). Maps of the apparent diffusion coefficient (ADC) were generated using a mono-exponential decay (all 4 b-values, or b-0 and 600 only). MT-ratio maps were generated from the MT data. Regions of interest (ROIs) on MRI scans were identified by consultant GI MRI radiologists. These regions were drawn on the DWI and MT raw images and then were copied on to the calculated maps. The ROIs were identified on Coronal fast spin echo (FSE) inversion recovery (IR) at 1.5T and T1 post contrast coronal images at 3T.
Median data of the individual pixel values from these regions were calculated. The T2-based Van Assche MRI score[5], C-reactive protein (CRP), PDAI score and fistula drainage are also assessed at each visit. The Van Assche score was measured from 1.5T images only, before the 3T images were viewed. Image quality at 3T provided clearer information about the size and position of the disease tissue. To determine how this difference in image information effected the ROIs drawn, volumes of the ROIs (both MT and DWI) were measured using Analyze 9 (Mayo Foundation, USA). As it was assumed that the 3T regions were more accurate the percentage (V1.5T/ V3T*100) and absolute differences (V3T-V1.5T) of the 1.5 T volumes compared to 3T volumes defined on the images were calculated for both DWI and MT as the images were acquired in different orientations. A response to treatment was defined as ≤4 of PDAI score. The study obtained REC approval (ref 16/EM/0433) and is registered in clinicaltrails.gov (ref: NCT03325582).
Results
After 12 weeks of treatment 5 of 9 patients were considered responders and 4 patients did not show a response to therapy (Fig 1).
Figure 1 also presents currently available data for PDAI, CRP and Van Assche score. The correlation coefficient of the Van Assche score with PDAI and CRP was r=.283 (P=.32) r=0.377 (P=0.18) respectively at baseline (N=14). Figure 3 presents the median DWI and MT ROI data for 9 subjects before and after commencing treatment at both field strength, showing responders and non-responders. Examples of MR images for a responder patient to treatment using 3T (Fig 4) and at 1.5T (Fig 5).
The median (interquartile range) baseline fistula volumes, were 13.8 mL (4.1-24.9 mL) for the 3T MT data and 7.8 mL (2.9-15.5 mL) for the DWI data. Median percentage change in defined fistula volume and absolute differences volumes (between 3T and 1.5 T data) measured across both visits were 123% (68-162%), -0.322ml (-2.0 – 3.6 mL) respectively for the MT data and 85% (46-109%), 0.28 mL (-0.36- 2.47 mL) for the DWI data. Therefore the 1.5T MT data tended to overestimate and the DWI data underestimate the fistula region.
Conclusion
The clinical and inflammatory markers did not correlate with the Van Assche score, this indicates that MRI is capturing different information about the disease activity. Volumes of the fistulas at the different field strength were not the same, which reflects both the different image quality available and sequences used. When the total cohort has been fully analysed it is envisaged that quantitative MRI scanning will add greater insight into perianal Crohn’s disease activity.Acknowledgements
Jazan University and Medical Research Council Confidence in Concept Funding, NIHR Nottingham Biomedical Research Centre.References
1. Association, A.G., American Gastroenterological Association medical position statement: perianal Crohn’s disease. Gastroenterology, 2003. 125(5): p. 1503-1507.
2. Rasul, I., et al., Clinical and radiological responses after infliximab treatment for perianal fistulizing Crohn's disease. The American journal of gastroenterology, 2004. 99(1): p. 82.
3. Siddiqui, M.R., et al., A diagnostic accuracy meta-analysis of endoanal ultrasound and MRI for perianal fistula assessment. Diseases of the Colon & Rectum, 2012. 55(5): p. 576-585.
4. Adler, J., et al., Magnetization transfer helps detect intestinal fibrosis in an animal model of Crohn disease. Radiology, 2011. 259(1): p. 127-135.
5. Van Assche, G., et al., Magnetic resonance imaging of the effects of infliximab on perianal fistulizing Crohn's disease. The American journal of gastroenterology, 2003. 98(2): p. 332-339.
Figures
