Patient Selection
This study was performed in compliance with the 1996 Health Information Portability and Accountability Act (HIPAA). Institutional review board granted approval and all patients provided written informed consent to participate in this prospective study. A standardized research protocol for the data collection was utilized.
A total of 156 consecutive patients were enrolled in this study. Inclusion criteria were informed consent, 18 years of age or older, known diagnosis of colonic CD, MRE performance, measurement of FCP levels within a maximum of two weeks prior to MRE, colonoscopy within a maximum of two weeks before or two weeks after the MRE, and no pharmacological therapy modification. Exclusion criteria were age younger than 18 years, no diagnosis of small bowel CD or small bowel and colonic CD confirmed by prior ileoscopy and biopsy, intolerance or contraindication to performance of MRE (such as pacemakers, MR-incompatible hardware, severe claustrophobia, and pregnancy), colonoscopy not performed two weeks before or two weeks after MRE, and FCP measurement not within a maximum of two weeks prior to MRE. The pharmacological therapies of patients were 5-aminosalicylic acids, corticosteroids, immunosuppressants, and/or biologic agents. Pharmacological therapy was not modified between FCP level measurement, MRE, and colonoscopy.
Crohn’s Disease Clinical Activity
At our institution, clinical disease activity was calculated for each patient at the time of MRE with the Crohn’s Disease Activity Index (CDAI) score. A CDAI score less than 150 indicated clinically inactive disease; scores of greater than 150 indicated active disease.
MR Enterography Acquisition
At our institution, MRE is performed using 3.0 Tesla magnet systems (Siemens Healthcare, Berlin, Germany). Patients ingest 1450 mL of a barium sulfate suspension (VoLumen; Bracco, Westbury, New York, U.S.A.) followed by 500 mL of water in divided doses one hour before the exam to achieve adequate bowel distention. Subsequently, a 0.5 mg dose of glucagon is administered intramuscularly prior to image acquisition to reduce bowel peristalsis. A second 0.5 mg dose of intramuscular glucagon (Eli Lilly, Indianapolis, Indiana, U.S.A.) is administered prior to the administration of intravenous gadolinium-based contrast material. Multi-planar MR imaging of the abdomen and pelvis was performed with a dedicated phased array torso coil using the following sequences (Table 1): coronal and axial T2-weighted half-Fourier acquisition single-shot turbo spin-echo (HASTE); T1-weighted dual gradient echo sequences; coronal fat-saturated T2-weighted true fast imaging with steady state precession (TrueFISP); coronal and axial T1 pre-contrast fat saturated (FS); serial dynamic coronal T1-weigted dynamic volume interpolated breath hold examination (VIBE) fat-saturated images obtained approximately 25 seconds, 60 seconds, 90 seconds after gadolinium contrast material (Gadavist 0.1mmol/kg; Bayer Healthcare Pharmaceuticals, Wayne, New Jersey, U.S.A.) intravenous injection; axial b50 and axial b800 diffusion-weighted images with apparent diffusion coefficient (ADC) mapping; and axial fat-saturated T1-weighted delayed VIBE images.
Table 1. MRE Acquisition Protocol
Sequence
|
Plane
|
Slice thickness/ Gap (mm)
|
TR/TE (ms)
|
FOV
|
Matrix
|
Flip Angle
|
T2W HASTE
|
Coronal
|
4/0
|
1500/120
|
440 x 440
|
384 x 224
|
90
|
T2W HASTE
|
Axial
|
5/1
|
1500/120
|
440 x 440
|
384 x 224
|
90
|
TrueFISP FS
|
Coronal
|
4/0
|
4.6/1.6
|
Variable
|
384 x 224
|
65
|
T1W dual
gradient echo
|
Axial
|
5/1
|
185/4.2 and 185/2.1
|
Variable
|
256 x 180
|
70
|
DWI (b0, 800)
|
Axial
|
5/1
|
TE minimum
|
Variable
|
132 x 132
|
90
|
T1W FS VIBE pre
|
Coronal
|
2/0
|
4.9/1.8
|
440 x 440
|
320 x 224
|
12
|
T1W FS VIBE pre
|
Axial
|
5/1
|
4.9/1.8
|
Variable
|
256 x 180
|
70
|
T1W FS VIBE post
|
Coronal
|
2/0
|
4.9/1.8
|
440 x 440
|
320 x 224
|
12
|
T1W FS VIBE post
|
Axial
|
5/1
|
4.9/1.8
|
Variable
|
256 x 180
|
70
|
MR Enterography Analysis: MaRIA
Two abdominal radiologists with twelve and five years of experience in interpreting MRE, respectively, independently reviewed the images from each MRE exam for the pattern and extent of abnormalities. The radiologists used Picture Archiving and Communication System (PACS) (Intellispace 4.4, Philips Healthcare, Amsterdam, Netherlands) on two separate workstations. The radiologists were blinded to colonoscopy results, FCP levels, and clinical and laboratory data.
Imaging features of inflammation are easily shown on MRE. Wall thickness that measures greater than 3 mm is abnormal; this thickening is due to edema and inflammation, which result in slightly increased signal intensity on T2-weighted HASTE and TrueFISP images. Contrast-enhanced T1-weighted VIBE images of the inflamed and thickened bowel show patterned wall hyperenhancement (12). Diffusion-weighted images show restricted diffusion in areas of active inflammation. Both T2-weighted images (HASTE and TrueFISP) and contrast-enhanced images show linear and transmural ulceration (12). Wall thickness, edema, contrast enhancement, and ulcers are the components used to calculate the MaRIA score of disease activity.
The radiologists each calculated a MaRIA scores for each MRE. The simplified (or segmental) MaRIA score for disease activity is calculated from the formula establish by Rimola et al.: (1.5 X wall thickness) + (0.02 X RCE enhancement) + (5 X edema) + (10 X ulceration) [5]. The cutoff value for active disease is ≥ 7 and for severe disease is ≥ 11. These cutoff values have shown high accuracy for diagnosis for both active disease: receiver operating characteristic (ROC) area 0.96, sensitivity 0.92, specificity 0.92; severe disease ROC area 0.93, sensitivity 0.87, and specificity 0.87, respectively. The radiologists arrived at a consensus if there was any discrepancy regarding the interpretation of the images by using the two most dominant of the four MRE features that comprised the MaRIA activity score formula.
Colonoscopic Analysis: CDEIS
At our institution, a pre-procedural oral preparation is used for bowel cleansing, which is essential for adequate mucosal examination. Variable types of sedation with cardiac and oxygen monitoring are for patient comfort and depend upon the predicted difficulty of the procedure. A digital rectal exam is performed to assess for skin tags, polyps, and fistula. Two gastroenterologists, with twelve and ten years of experience with colonoscopy of patients with CD, used a high-definition colonoscope (EC-3890Li, RiCoh, Tokyo, Japan) that is introduced through the anus and advanced through the colon. The colonoscope is connected to a multichannel system for air insufflation, suction, water, video monitor, and power supply.
The gastroenterologists were aware of the patient’s diagnosis of CD but blinded to MRE results. The gastroenterologists calculated total CDEIS score for each patient by assessing for deep ulceration (no = 0, yes = 12), superficial ulceration (no = 0, yes = 6), surface involved by disease (0–10), ulcerated surface (0–10), and ulcerated or non-ulcerated stenosis (no = 0, yes = 3).
Fecal Calprotectin Analysis
The concentration of fecal calprotectin was measured at an outside laboratory using the values recommended by the laboratory (Quest Diagnostics, Secaucus, New Jersey, U.S.A.) using patients’ fecal samples (40-100 mg) and a polyclonal antibody quantitative enzyme-linked immunoassay. The concentration of calprotectin in the fecal sample was calculated using the values recommended by the laboratory. Results were expressed in microgram per gram of feces. The analytical sensitivity is 6.25 ug/mL. In this laboratory, values above 250 mg/g are considered abnormally elevated and values below 50 mg/g are considered normal.
Statistics
The Kruskal-Wallis test was performed to evaluate the association between FCP levels and MaRIA, FCP and CDEIS, and CDEIS and MaRIA. Specifically, the Kruskal-Wallis test was used to assess difference for these non-normally distributed variables. We performed multiple univariate analyses between FCP and colonoscopy, FCP and MRE, and MRE and colonoscopy and then multivariate analysis between FCP, colonoscopy, and MRE to assess if there is independent positive correlation between each pair of these biometric tests and then between all three biometric tests.
We had no missing information for the data presented in this study. We used a correlative approach and applied multivariate regression analysis to evaluate the effect of CDEIS and MaRIA on FCP levels. All statistical analysis was performed using SAS software, version 9.4 (SAS Institute Incorporated, Cary, North Carolina, U.S.A.). A probability value of P < 0.05 was considered to be statistically significant.