List of Abbreviations
Apparent diffusion coefficient
The Crohn’s Disease MRI Index
Contrast enhanced ultrasonography
Computed tomography enterography
The European Crohn’s and Colitis Organisation
The European Society of Gastrointestinal and Abdominal Radiology
Half-Fourier acquisition single-shot turbo spin-echo
Inflammatory bowel disease
The Magnetic Resonance Index of Activity
Maximum intensity projection
Magnetic resonance enterography
Small intestine contrast ultrasonography
Single-shot fast spin-echo
Inflammatory bowel disease (IBD) encompasses both Crohn’s disease (CD) and ulcerative colitis (UC). CD is a chronic immune-mediated transmural inflammatory disorder of the gastrointestinal tract that is characterized by a progressive and destructive course, leading to irreversible structural bowel damage. In population-based cohorts, up to one-third of patients with CD have evidence of bowel damage at the time of diagnosis (stricturing or penetrating complications), often underdiagnosed without the aid of cross-sectional imaging. Intestinal imaging has revolutionized diagnostic and treatment algorithms in CD patients. Cross-sectional imaging techniques are now viewed as complementary to ileocolonoscopy, providing assessments of inflammation in regions inaccessible to standard endoscopic techniques or isolated intramural disease. Applications include evaluating disease extent and severity, differentiating CD from UC, detecting disease complications (strictures, fistulas, and/or abscesses), assessing response to medical therapy, and detecting postoperative recurrence.
Cross-sectional imaging techniques using enterography protocols with either computed tomography enterography (CTE) or magnetic resonance enterography (MRE) and small intestine ultrasonography can identify and quantify transmural structural damage and disease activity. This chapter explores these IBD imaging modalities followed by a discussion of their applications for the diagnosis of stricturing and penetrating disease, as well postoperative complications.
Computed Tomography Enterography
CTE technique is tailored to maximize small bowel wall assessment. Adequate technique requires that after a 4-hour fast, the patient ingests a large volume of neutral enteric contrast material that contains sugar alcohols or osmotic laxatives, agents that fill the bowel lumen and prevent absorption of water and ultimately provide optimal bowel distension. Nearly 1300–1800 cc of oral contrast material is ingested over 30–60 min. These neutral enteric contrast agents have the attenuation of water and thus increase the conspicuity of the enhancement of actively inflamed bowel wall following the administration of intravenous (IV) contrast material while positive oral contrast material such as is often used for routine CT abdomen and pelvis can obscure pathologic mural enhancement. Additional differences from a routine CT abdomen and pelvis include images obtained at 45 s after infusion of contrast material begins (enteric phase), use of thinner axial images, sagittal and coronal reformations, maximum intensity projection images, and extension to the perineum (alerts to perianal disease). CTE has a high sensitivity (>90%) and specificity (∼90%) for the detection of small bowel CD, and it has been validated against clinical, histologic, and endoscopic assessments. The diagnostic accuracy in patients with suspected or established CD is comparable to that of MRE, with CTE reported to have higher and consistent image quality. In addition to accurate disease detection in patients with suspected CD, CTE has been shown to play a vital role in the management of patients with known CD as it can detect clinically occult inflammation and can detect and help map out complications such as fistulas, abscesses, and strictures. Ultimately the findings at CTE frequently alter treatment plans and impact corticosteroid usage.
Magnetic Resonance Enterography
Standard CD MRE technique is less standardized across the country than CTE, but in general the most useful sequences include unenhanced T2-weighted images and gadolinium-enhanced T1-weighted sequences in multiple planes of acquisition. Similar to CTE, patients are asked to ingest a large-volume oral agent, and IV contrast is utilized. Post-gadolinium images are obtained 45 s after injection followed by dynamic images. Spasmolytics are useful for reducing bowel peristalsis and motion artifacts. The most frequently used oral contrast agents for MRE are considered “biphasic”, that is, they are low signal intensity on T1-weighted images (thus increasing conspicuity of inflammation following administration of contrast material) and they are hyperintense on T2-weighted images, allowing for better assessment of the bowel wall thickness. The T2-weighted images are often obtained both with and without fat saturation; those with fat saturation allow for better detection of high T2 signal mural and perienteric inflammation/edema, and those without fat saturation help with accurate bowel wall assessment and allow better visualization of the vasa recta. Images are initially obtained during the enteric phase at 45 s after the initiation of the contrast material infusion. Because of the lack of ionizing radiation, many sequential acquisitions can be obtained before and after injection followed by dynamic images. Delayed imaging at 7 min may allow differentiation of inflammatory and fibrotic components of strictures. Gadolinium-enhanced T1-weighted images allow interrogation of bowel wall enhancement characteristics, transmural ulcers, fistulas, sinus tracts, comb sign (dilated vasa recta), and perienteric abnormalities, whereas bowel wall thickening and mural edema are better evaluated in the T2-weighted images. MRE, similar to CTE, has high sensitivity (>90%) and specificity (>90%) for the diagnosis of active small bowel inflammation.
MRE protocols may now also include diffusion-weighted images (DWIs). This technique provides quantifiable information that may complement T1/T2-weighted images and can be acquired without IV contrast material. High signal intensity on DWIs and corresponding decrease in the apparent diffusion coefficient (ADC) suggest restricted diffusion of water molecules due to active inflammation. A prospective, noninferiority study in 44 patients with known or suspected CD compared precontrast sequences alone with DWIs to conventional MRE with gadolinium contrast. In this study, precontrast sequences with DWIs identified active inflammation in the terminal ileum with a sensitivity and specificity of 93% and 67%, respectively. DWIs can be considered in CD patients who require serial magnetic resonance (MR) assessments, are pregnant, or have impaired renal function and therefore a desire to limit gadolinium exposure.
MRE scoring systems have been developed to allow quantitative assessments of lesions for target-directed medical therapy in clinical practice and trials. These have been developed in comparison to an external reference of either ileocolonoscopy (Magnetic Resonance Index of Activity [MaRIA] score and Nancy score) or histopathology of a resected specimen (Crohn’s disease MRI Index [CDMI] score). Of these, the MaRIA and CDMI have been validated, with only the MaRIA score shown to be responsive and reliable in assessing the response to therapy in patients with CD. An MRE-based scoring system utilizing DWI has also been developed, the Clermont score, which can be obtained without bowel preparation or colonic enema. Finally, recent efforts have also centered on the development of a scoring system that accounts for clinical, endoscopic, and radiological information to assess the burden of disease, the Lémann index. This index accounts for previous operations, strictures, and penetrating lesions across the entire gastrointestinal tract. This system will likely need to be simplified before widespread use in clinical practice if feasible and practical.
Ultrasonography (US) in IBD has been extensively studied in Europe and selective North American IBD academic centers. The unenhanced gray-scale US protocol is often performed as a two-step process with 3–8 mHz probes followed by higher frequency linear probes (7–9 mHz). The lower frequency probes provide a better overview of the abdomen and can sometimes more completely image larger lesions in the intestine. The curved or linear higher frequency transducers offer a more detailed transabdominal evaluation of the small intestine, allowing visualization of the bowel wall layers. A common imaging approach is to examine the entire colon in a retrograde fashion from the rectum to the cecum, followed by evaluation of the distal ileum, and subsequent systematic surveillance of the remainder of the small bowel in all four quadrants of the abdomen. If pathology is detected, bowel wall thickness should be documented, as well as presence or absence of mural stratification, luminal narrowing/stenosis, and/or bowel dilatation. In addition, the motility pattern and the presence or absence of fibrofatty proliferation, and perienteric lymph nodes should be noted. The pooled per-patient sensitivity and specificity for the diagnosis of CD are 85% (95% confidence interval [CI], 83%–87%) and 98% (95% CI, 95%–99%), respectively, at IBD centers with experience with bowel US, with highest accuracy at the terminal ileum and left hemicolon. Sonographic technology has since evolved with the expanded use of Doppler ultrasound. This tool quantitates the amount of blood flow in the wall of the bowel and surrounding mesentery by the absolute velocity of flow and density of mural blood vessels. US measurements have been used to assess response to medical treatment.
Contrast enhanced ultrasonography (CEUS) protocols contain the intravenous administration of contrast containing microbubbles filled with sulfur hexafluoride (SonoVue and Lumason, SV, Bracco, Italy), octafluoropropane (Definity and Luminity, Lantheus Medical Imaging, USA), perflutren (Optison, GE Healthcare, USA), or perfluorobutane (Sonazoid, GE Healthcare, USA). The contrast agents demonstrate tissue perfusion with time blood-pool imaging. These agents have not yet been approved by the Food and Drug Administration in the United States for use in IBD patients. CEUS has been shown to perform better than gray-scale US or color Doppler at detecting active CD, with 93.5% sensitivity, 93.7% specificity, and 93.6% overall accuracy. A recent meta-analysis of eight studies utilizing CEUS for the detection of active CD demonstrated a pooled sensitivity of 0.94 (95% CI 0.87–0.97) and specificity of 0.79 (95% CI 0.67–0.88).
Small intestine contrast ultrasonography (SICUS) is another modification of unenhanced gray-scale US involving the ingestion of oral contrast material (usually 250–800 mL of polyethylene glycol), after an overnight fast. This has been shown to improve the detection of proximal small bowel inflammatory lesions and strictures compared to conventional US. A scoring system, the quantitative sonographic lesion index for CD has been developed, that has been shown to be responsive to medical therapy with anti-TNF-α agents.
Comparison of Imaging Modalities
Understanding the advantages and disadvantages of each imaging modality is a critical step to provide evidence-based patient-specific medical care. When comparing CTE to MRE, CTE has widespread availability and rapid scan times, making it the preferred imaging option in the acute setting. Furthermore, rapid image acquisition obviates the need for an antiperistaltic agent (such as glucagon) that may lead to patient discomfort during an MRE, and it allows for imaging of patients who are frail, claustrophobic, unable to lie flat for long periods and/or reliably hold their breath. The main limitation of CTE is the adverse health risks related to radiation exposure with serial imaging in young patients and IV contrast load in pregnant patients or patients with allergy or with renal insufficiency. In contrast, MRE avoids radiation exposure, is able to study temporal changes in bowel over the multiple sequences (confirm findings or refute suspicion of findings if they normalize), and is able to provide additional information regarding small bowel motility and potential strictures (dynamic imaging). Limitations of MRE include longer acquisition time, cost, availability, variability (both on part of technique and patient ability—breath holding and staying still), worse spatial resolution (but better tissue contrast resolution) than CT, claustrophobia or patients with pacemaker or other implants that may preclude an MRE, more artifacts with gas, and bowel motion. The European Society of Gastrointestinal and Abdominal Radiology (ESGAR)/European Crohn’s and Colitis Organisation (ECCO) consensus statement on imaging in IBD has expressed preference for MRE over CTE in most non-acute settings.
Advantages of US compared to MRE and CTE include its portability, widespread availability, improved patient tolerance (no oral contrast and shorter examination times), and low cost. It also represents an additional radiation-free alternative to CTE. It is, however, more operator dependent than MR or CT with variable image quality and limited experience in IBD imaging, limited by the amount of intestinal gas and has limited utility in heavier patients with deep lying segments of bowel. Some institutions may utilize endovaginal imaging of deep pelvic loops to overcome this, though limited by patient tolerance. Unenhanced US compared with MRE (ileocolonoscopy as gold standard) has been shown to be less accurate than MRE in defining the extent of CD, especially with CD involvement of greater than 30 cm of small bowel as well in the detection of enteroenteric fistulas. SICUS compared to a CT enteroclysis (protocol similar to CTE, with oral contrast instilled via a jejunal enteric tube to achieve more robust distension) performed 3 months apart was shown to have comparable performance for small bowel disease extent, presence of strictures, abscesses, and fistulae. SICUS and MRE have also been shown to be comparable in detecting small bowel complications of CD when correlated to the intraoperative findings.
In a population-based cohort of CD patients, 81.4% of the patients had non-stricturing non-penetrating disease, 4.6% had stricturing disease, and 14.0% had penetrating disease. The cumulative risk of developing either complication increased over time: 18.6% at 90 days, 22.0% at 1 year, 33.7% at 5 years, and 50.8% at 20 years after diagnosis. One proposed imaging definition of stricturing disease is luminal narrowing in an area of CD with unequivocal proximal dilation (upstream lumen ≥3 cm) or persistence over two time points (e.g., multiple pulse sequences or delayed series). Regardless of the definition, symptoms of nausea, vomiting, bloating after consumption of higher residue foods, and physical examination findings of a succussion splash or abdominal distension should prompt further assessment with cross-sectional imaging.
CTE has been compared to ileoscopy and histopathology for the detection of small bowel stenosis with a sensitivity of 85%–93% and specificity of 100%. While the accuracy for a single stricture may approach 100%, the accuracy for the number of strictures is only 83%, and CTE may exceed or underestimate the extent of complicated disease in 31% of CD patients. Although the CTE findings of mesenteric hypervascularity (dilated vasa recta or the “comb sign”), mural hyperenhancement and/or stratification, and perienteric mesenteric edema predict tissue inflammation, the absence of these findings does not predict tissue fibrosis. ECCO recommends that in patients with established CD of the small bowel, small bowel strictures must first be excluded by clinical history and radiographic imaging before obtaining small bowel capsule endoscopy.
Robust data also exist for the use of MRE to diagnose small bowel strictures ( Fig. 7.1 ). Similar to CTE, MRE has sensitivity for detection of strictures ranging from 75% to 100% with a specificity of 91% to 100%. In terms of specific imaging findings, MRE findings of bowel wall thickening have been shown to have a negative association with response to medical therapy and a positive association with small bowel fibrosis. In a pediatric cohort of CD patients with symptomatic strictures who underwent preoperative MRE with histopathologic correlation, prestenotic upstream small bowel dilatation greater than 3 cm was significantly associated with confluent transmural fibrosis. A recent study correlated the percentage of gadolinium enhancement gain between 70 s and 7 min and the pattern of enhancement at 7 min with the degree of fibrosis ( P < .01 for both). Using percentage of enhancement gain, MRE was able to discriminate between mild–moderate and severe fibrosis deposition with a sensitivity of 0.94 and a specificity of 0.89. Positron emission tomography (PET)/MR is an emerging tool to differentiate inflammatory from fibrotic strictures using a combination imaging biomarker of ADC and PET maximum standardized uptake value, with a cutoff less than 3000 associated with an accuracy, sensitivity, and specificity values of 71%, 67%, and 73%, respectively.
Small bowel ultrasound can also identify intestinal complications of CD. Bowel strictures may appear at US as intestinal segments with thickened walls, narrowed lumen, with or without associated upstream bowel dilatation (i.e., diameter 25–30 mm), with increased fluid content and hyperperistalsis or stagnation. Furthermore, loss of stratification with increased mural vascularity or perienteric vascularity and large adjacent lymph nodes would raise suspicion for active inflammatory stricture, whereas the presence of stratification may indicate a higher degree of fibrosis. Across ten studies, including eight that used surgical specimens as a reference standard, the sensitivity and specificity of bowel US for assessing stricturing complications were 79.7% (95% CI, 75.2%–84.2%) and 94.7% (95% CI, 89.7%–99.8%), respectively. SICUS protocol has been shown to improve the ability of bowel US in detecting strictures while CEUS with qualitative parameters analysis seems to be a promising tool for distinguishing inflammatory from fibrostenotic lesions in CD patients. An algorithmic approach to diagnosis of stricturing CD is represented in Fig. 7.2 .
Putting It All Together: Strictures
MRE is the preferred imaging strategy for stricturing CD in the outpatient setting. MRE is also the imaging study of choice in IBD patients with renal insufficiency and in pregnant patients, using a noncontrast protocol with T2-weighted images and DWIs. CTE is often preferred in inpatients due to its faster image acquisition, in nonpregnant outpatients with claustrophobia, and in patients with implanted medical devices (contraindicated for MR). Bowel US using either the CEUS or SICUS protocol can be utilized in select centers with expertise in using this modality for IBD, especially in disease limited to the terminal ileum, pediatric, and pregnant patients.
Intra-abdominal Penetrating Disease
Penetrating complications of CD include sinus tracts, fistulas, inflammatory masses, perforation, and abscesses. Sinus tracts are blind ending, extending to the mesentery, fascial planes, or longitudinally within the bowel wall. Intra-abdominal fistulas connect two epithelialized structures. Simple abdominal fistulas appear as a solitary extraenteric soft tissue tract, with or without internal air or fluid and with associated angulated or tethered bowel loops. Complex fistulas are multiple and/or branching. When severe, a complex fistula may have an asterisk-shaped network of inflammatory tracts, often associated with an interloop abscess or inflammatory mass, and with associated angulated or tethered bowel loops ( Fig. 7.3 ). Fistulas may extend to adjacent bowel, bladder, urethra or female-reproductive organs, or skin ( Figs. 7.4A,B and 7.5 ). An inflammatory mass (now favored over the term “phlegmon”) describes an area of dense mesenteric inflammation adjacent to severe mural inflammation or penetrating complications in the absence of an organized abscess.