Celiac disease is a common inflammatory disease of the small intestine triggered by gluten in genetically susceptible individuals. Diagnosis is made by serologic testing and upper endoscopy with small bowel biopsy in most individuals. Celiac patients may present with abdominal pain or nonspecific gastrointestinal complaints that result in radiologic imaging before diagnosis of celiac disease. Wireless video capsule endoscopy, device-assisted enteroscopy, and enterography allow careful examination of the entire small bowel and targeted sampling of suspicious lesions. This review focuses on the role of device-assisted enteroscopy and radiologic imaging, in particular enterography, in celiac disease.
- •
Recent advances in small bowel imaging technologies have improved the care of patients with small bowel diseases. Small bowel endoscopic and radiologic technologies are complementary and are often used in conjunction.
- •
In patients with celiac disease and gastrointestinal symptoms, radiologic imaging may be diagnostic of celiac disease.
- •
It is critical that radiologists and gastroenterologists are familiar with findings suggestive of celiac disease with new imaging modalities.
- •
Video capsule endoscopy, enterography, and device-assisted enteroscopy are usually reserved for those with alarm symptoms, refractory celiac disease, or suspicion of small bowel lymphoma/adenocarcioma.
Introduction
The small intestine is the longest organ of the gastrointestinal tract, which is fanned on the mesenteric stalk in the abdominal cavity. The multiple folds of small bowel loops in the abdominal cavity and peristalsis make it difficult to examine using standard endoscopic and radiologic imaging techniques. As a result, the small intestine has long been considered the black hole of the gastrointestinal tract. Upper endoscopy, push enteroscopy, and colonoscopy allow examination of only a small portion of the jejunum and distal ileum. Radiologic imaging by small bowel barium study and traditional abdominal computed tomography (CT) show luminal findings suggestive of celiac disease but provide poor examination of the small bowel wall. In addition, small bowel series is a time-consuming, operator-dependent study that has fallen out of favor for CT imaging. Recent advances in endoscopic imaging technologies (capsule and device-assisted eneroscopy) have enabled detailed visualization of the entire small bowel mucosa. Radiologic advances (CT enterography [CTE] and magnetic resonance enterography [MRE]) have markedly improved examination of the small bowel wall and surrounding structures. Intraoperative enteroscopy, considered the gold standard of complete enteroscopy, has largely been replaced. New imaging technologies provide less-invasive methods, with excellent reader agreement, for examining the small bowel. These technologies have improved the diagnosis and management of patients with small bowel diseases, such as small bowel bleeding, Crohn, disease, and celiac disease.
Celiac disease is a common inflammatory disease of the small bowel that affects 1% of the white population. It is triggered, in genetically predisposed individuals, by the ingestion of gluten, a protein component of wheat, rye, and barley. Most individuals are diagnosed with celiac disease in adult life. Celiac disease may present with classic (diarrhea-predominant) symptoms or atypical symptoms or may be asymptomatic and detected via screening. Atypical presentations of celiac disease include nonspecific abdominal pain/dyspepsia, constipation, bloating, reflux, infertility, anemia, osteoporosis, dental enamel defects, short stature, vitamin deficiencies, fatigue, or neurologic problems, such as neuropathy or ataxia. The rash of dermatitis herpetiformis is virtually always associated with celiac disease whereas intestinal biopsy may be normal with gluten ataxia. Serology for the antibodies directed against tissue transglutaminase (tTG IgA) is the best screening test for celiac disease. The gold standard for diagnosis is upper endoscopy with small bowel biopsy. Patients with positive serology should have an upper endoscopy performed with 4 to 6 biopsies of the small intestine with samples from both the bulb and the second portion of the duodenum to maximize yield. If the suspicion for celiac disease is high, endoscopy with biopsy should be performed despite negative serology results. Endoscopic findings of celiac disease include loss of folds and scalloping, a mosaic pattern, and fissuring of mucosa ( Fig. 1 ). Endoscopic abnormalities are not present in all cases and biopsies should be obtained even if the duodenum appears normal. Marsh and Oberhuber and colleagues described the histologic changes of celiac disease as increased intraepithelial lymphocytes, crypt hyperplasia, and villous atrophy. These findings largely account for abnormalities detected on radiologic imaging studies. Endoscopic biopsies of the duodenum may be normal if the disease is patchy. Jejunal biopsies may improve diagnostic yield in such patients.
The only current available treatment of celiac disease is a gluten-free diet. Up to 30% of celiac patients experience continued symptoms on a gluten-free diet or incomplete histologic recovery and are considered nonresponsive. Other causes of ongoing symptoms include continued gluten exposure, microscopic colitis, small intestinal bacterial overgrowth, lactose or fructose intolerance, pancreatic exocrine insufficiency, and refractory celiac disease (RCD). RCD is defined as persistent diarrhea with villous atrophy, crypt hyperplasia, and inflammation, despite adherence to a strict gluten-free diet for 6 to 12 months. Its prevalence is unknown. RCD is classified into type 1 and type 2 depending on intraepithelial lymphocyte expression. In RCD type 1, intraepithelial lymphocytes have normal surface expression of both CD3 and CD8 with a polyclonal T-cell receptor. In RCD type 2, however, an abnormal lymphocyte population is present with a loss of surface CD3 and CD8 expression, retention of intracellular CD3, and a monoclonal T-cell receptor rearrangement. In a study of 57 patients with RCD, the 5-year survival rate was 93% in patients with RCD type 1 and 44% with RCD type 2. Enteropathy-associated T-cell lymphoma (EATL) occurs in more than 50% of patients with RCD type 2 and is a significant cause of mortality. Evaluation of poorly responsive and refractory celiac patients includes consultation with a skilled dietician, endoscopy with biopsy, stool studies, hydrogen breath testing, and, in cases of alarm symptoms and refractory disease, video capsule endoscopy, enterography, and deep enteroscopy if a suspicious lesion is identified. This review focuses on the role of standard radiologic imaging, enterography, and device-assisted enteroscopy in celiac disease.
Device-assisted enteroscopy
Device-assisted enteroscopy allows for diagnosis, tissue sampling, and therapy. The procedure can be performed using balloons or a spiral element attached to the endoscope. Therapeutic options can be performed during enteroscopy, such as hemostasis, lesion marking, and stricture dilation/stenting. The majority of published literature has involved double-balloon enteroscopy (DBE) because it has been available for the longest period of time.
Balloon-Assisted Enteroscopy
DBE, or push-and-pull enteroscopy, was introduced in 2003 and allows complete examination of the small bowel via the oral (antegrade) or rectal (retrograde) approach or both approaches. The DBE system (Fujinon, Wayne, New Jersey) consists of a 140-cm polyurethane overtube back-loaded on a 200-cm enteroscope with 2 latex balloons attached at the tips of the overtube and enteroscope. Single-balloon enteroscopy (SBE) (Olympus Optical, Tokyo, Japan) uses a stiffer enteroscope and single, latex-free, balloon on the end of the overtube. A series of push-and-pull maneuvers are performed with serial inflation and deflation of the balloons. These maneuvers hold the bowel in place, prevent looping, and allow advancement deep into the small bowel by pleating the small bowel onto the overtube. The exact depth of insertion is difficult to measure. The procedure is time consuming and the average procedure time of DBE in initial experience ranged from 70 to 110 minutes. The retrograde approach is particularly challenging due to retroflexion of the thin flexible endoscope in the cecum on advancement into the ileocecal valve with unstable intubation of the terminal ileum. In several initial studies, the overall diagnostic yield of DBE ranged from 43% to 80% and total enteroscopy varied from 5% to 86%. Success of total enterosocpy is influenced by endoscopist experience, procedure time, patient body habitus, and prior abdominal surgery. Complication rates for DBE in an international multicenter study were 0.8% and 4.3% for diagnostic and therapeutic procedures, respectively. Reported complications of DBE are pancreatitis, bleeding, and perforation. Perforation was highest in the setting of ulcerating disease and altered small bowel anatomy. The absolute contraindications to balloon-assisted enteroscopy is small bowel obstruction, and in the case of DBE, latex allergy with relative contraindications being coagulopathy, pancreatitis, and large esophageal varices. When compared with push enteroscopy, DBE had higher diagnostic yield (63% vs 44%) and greater depth of insertion (230 cm vs 80 cm). In 1 study, carbon dioxide insufflation rather than air, increased the depth of insertion and reduced patient discomfort.
In small studies comparing the DBE and SBE techniques, diagnostic yields were similar, but total enteroscopy rate was higher with DBE. In a prospective, multicenter trial, the DBE system had a significantly higher diagnostic yield and total enteroscopy rate when 2 balloons were used compared with a single balloon used at the tip of the overtube. This study did not use the actual SBE system, which contains a stiffer enteroscope. It is unclear if using the SBE system would have changed these results.
The literature on the use of enteroscopy in celiac disease is limited. A small series using DBE for the evaluation of malabsorption suggests DBE was useful in excluding complications of celiac disease ( Figs. 2 and 3 ). Because balloon-assisted enteroscopy is time consuming and invasive, it best complements video capsule endoscopy and enterography when suspicious lesions are identified. In a series of 21 patients with RCD, DBE was useful in detecting or excluding complications, such as EATL or ulcerative jejunitis. In this study, low-risk lesions were defined as a reduction (<3 per endoscopic field of view) or loss of folds, scalloping, nodularity, and presence of visible vessels after air insufflations. High-risk lesions were defined as stenosis and ulcerations more than 5 mm in diameter. Hadithi and colleagues detected EATL in 5 of 21 patients (24%) that appeared as circumferential, discrete, or confluent ulcerations whereas 2 of 21 patients (9%) were diagnosed with ulcerative jejunitis in the absence of EATL. DBE helped to exclude EATL in 4 patients with CT findings suggestive of malignancy due to small bowel wall thickening. A recent small case series examined the role of DBE using virtual chromoendoscopy (Fujinon intelligent color enhancement [FICE]). There was no improvement in the detection of features of celiac disease using FICE.
Spiral Enteroscopy
Spiral enteroscopy (SE) is a method of device-assisted enteroscopy that uses rotational energy to pleat the small bowel. The Endo-Ease Discovery SB (Olympus Optical), a 118-cm polyvinyl chloride overtube with a 21-cm spiral element at the tip, can be back-loaded and locked onto an enteroscope. Using clockwise rotation, the device pleats the small bowel on to the overtube until the device can no longer be advanced. The enteroscope is subsequently unlocked and advanced through the overtube with a subsequent series of hook-and-suction maneuvers to further pleat small bowel. Advantages of SE include reduced procedure times and stability for therapeutics deep in the small intestine, because the enteroscope can be removed and reinserted through the overtube. In the initial experience, mean procedure time was 36.5 minutes and diagnostic yield 33%. In a subsequent prospective multicenter study of SE, the diagnostic yield was 65% and average procedure time 45 minutes. In a small, prospective, crossover, single-center study comparing SE and DBE, SE had a reduced examination time (43 minutes vs 65 minutes) but decreased insertion depth compared with DBE (250 cm vs 310 cm). Although there are no published studies of SE in celiac disease, the device has been used in celiac patients without complications (personal experience).
Device-assisted enteroscopy
Device-assisted enteroscopy allows for diagnosis, tissue sampling, and therapy. The procedure can be performed using balloons or a spiral element attached to the endoscope. Therapeutic options can be performed during enteroscopy, such as hemostasis, lesion marking, and stricture dilation/stenting. The majority of published literature has involved double-balloon enteroscopy (DBE) because it has been available for the longest period of time.
Balloon-Assisted Enteroscopy
DBE, or push-and-pull enteroscopy, was introduced in 2003 and allows complete examination of the small bowel via the oral (antegrade) or rectal (retrograde) approach or both approaches. The DBE system (Fujinon, Wayne, New Jersey) consists of a 140-cm polyurethane overtube back-loaded on a 200-cm enteroscope with 2 latex balloons attached at the tips of the overtube and enteroscope. Single-balloon enteroscopy (SBE) (Olympus Optical, Tokyo, Japan) uses a stiffer enteroscope and single, latex-free, balloon on the end of the overtube. A series of push-and-pull maneuvers are performed with serial inflation and deflation of the balloons. These maneuvers hold the bowel in place, prevent looping, and allow advancement deep into the small bowel by pleating the small bowel onto the overtube. The exact depth of insertion is difficult to measure. The procedure is time consuming and the average procedure time of DBE in initial experience ranged from 70 to 110 minutes. The retrograde approach is particularly challenging due to retroflexion of the thin flexible endoscope in the cecum on advancement into the ileocecal valve with unstable intubation of the terminal ileum. In several initial studies, the overall diagnostic yield of DBE ranged from 43% to 80% and total enteroscopy varied from 5% to 86%. Success of total enterosocpy is influenced by endoscopist experience, procedure time, patient body habitus, and prior abdominal surgery. Complication rates for DBE in an international multicenter study were 0.8% and 4.3% for diagnostic and therapeutic procedures, respectively. Reported complications of DBE are pancreatitis, bleeding, and perforation. Perforation was highest in the setting of ulcerating disease and altered small bowel anatomy. The absolute contraindications to balloon-assisted enteroscopy is small bowel obstruction, and in the case of DBE, latex allergy with relative contraindications being coagulopathy, pancreatitis, and large esophageal varices. When compared with push enteroscopy, DBE had higher diagnostic yield (63% vs 44%) and greater depth of insertion (230 cm vs 80 cm). In 1 study, carbon dioxide insufflation rather than air, increased the depth of insertion and reduced patient discomfort.
In small studies comparing the DBE and SBE techniques, diagnostic yields were similar, but total enteroscopy rate was higher with DBE. In a prospective, multicenter trial, the DBE system had a significantly higher diagnostic yield and total enteroscopy rate when 2 balloons were used compared with a single balloon used at the tip of the overtube. This study did not use the actual SBE system, which contains a stiffer enteroscope. It is unclear if using the SBE system would have changed these results.
The literature on the use of enteroscopy in celiac disease is limited. A small series using DBE for the evaluation of malabsorption suggests DBE was useful in excluding complications of celiac disease ( Figs. 2 and 3 ). Because balloon-assisted enteroscopy is time consuming and invasive, it best complements video capsule endoscopy and enterography when suspicious lesions are identified. In a series of 21 patients with RCD, DBE was useful in detecting or excluding complications, such as EATL or ulcerative jejunitis. In this study, low-risk lesions were defined as a reduction (<3 per endoscopic field of view) or loss of folds, scalloping, nodularity, and presence of visible vessels after air insufflations. High-risk lesions were defined as stenosis and ulcerations more than 5 mm in diameter. Hadithi and colleagues detected EATL in 5 of 21 patients (24%) that appeared as circumferential, discrete, or confluent ulcerations whereas 2 of 21 patients (9%) were diagnosed with ulcerative jejunitis in the absence of EATL. DBE helped to exclude EATL in 4 patients with CT findings suggestive of malignancy due to small bowel wall thickening. A recent small case series examined the role of DBE using virtual chromoendoscopy (Fujinon intelligent color enhancement [FICE]). There was no improvement in the detection of features of celiac disease using FICE.
