Pediatric Radiology

, Volume 38, Issue 5, pp 497–510

Fluoroscopic and CT enteroclysis in children: initial experience, technical feasibility, and utility

Authors

  • Shanaree Brown
    • Indiana University School of Medicine
    • Department of RadiologyUniversity of Missouri
    • Section of Pediatric RadiologyRiley Hospital for Children
    • Department of Radiology, Indiana University School of MedicineRiley Hospital for Children
  • Kumar Sandrasegaran
    • Department of Radiology, Indiana University School of MedicineRiley Hospital for Children
  • S. Greg Jennings
    • Department of Radiology, Indiana University School of MedicineRiley Hospital for Children
  • Joshua Garrett
    • Department of Radiology, Indiana University School of MedicineRiley Hospital for Children
  • Arunan Skantharajah
    • Indiana University
  • Dean T. Maglinte
    • Department of Radiology, Indiana University School of MedicineRiley Hospital for Children
Original Article

DOI: 10.1007/s00247-008-0754-8

Cite this article as:
Brown, S., Applegate, K.E., Sandrasegaran, K. et al. Pediatr Radiol (2008) 38: 497. doi:10.1007/s00247-008-0754-8

Abstract

Background

Partial small-bowel obstruction can be difficult to diagnose on clinical examination. These obstructions might not be detected on routine abdominal/pelvic CT.

Objective

To evaluate the feasibility, safety, and techniques of fluoroscopic enteroclysis (FE) and CT enteroclysis (CTE), and to review their indications and findings in children.

Materials and methods

We retrospectively reviewed all enteroclysis studies in children younger than 18 years performed between January 2002 and March 2007. We correlated the results with other abdominal imaging and surgical and pathological findings.

Results

The review revealed 112 FE and 74 CTE studies performed in 175 children (mean age 14 years, range 3–18 years). FE and CTE studies were performed most commonly for evaluation of known Crohn disease (FE 38%, CTE 29%) and abdominal pain (FE 26%, CTE 26%). One FE study was terminated because of patient anxiety, and one CTE study was terminated because of patient discomfort. No complications of FE or CTE were reported. The findings were normal in 54% of the FE studies and 46% of the CTE studies. The most common small bowel diagnoses were Crohn disease (FE 34%, CTE 28%) and partial small bowel obstruction (FE 3%, CTE 10%). Two FE studies (2%) and 14 CTE studies (19%) showed abnormalities outside the small bowel. In 54 patients with inflammatory bowel disease, 11 FE studies and 25 CTE studies showed additional bowel abnormalities. Overall, 14 and 21 patients had surgery as a result of the findings of FE and CTE, respectively.

Conclusion

FE and CTE are safe, feasible, and accurate in depicting small-bowel pathology in children. These techniques can be particularly useful in children with Crohn disease involving the small bowel.

Keywords

Fluoroscopic enteroclysisCT enteroclysisChildrenCrohn diseaseSmall bowel obstruction

Introduction

Radiological evaluation of the small bowel can be challenging because of its inaccessibility to endoscopy and its length and redundancy on imaging. Small-bowel follow-through (SBFT) has been the most commonly performed method of investigation because of its ease of use, availability, and low cost, but recently several other methods of small-bowel investigation have been developed.

Abdominal multidetector CT (MDCT) is fast and accessible, and provides images of consistently high quality. Not surprisingly, it is increasingly used to evaluate both intestinal and extraintestinal pathology. Both children and adults, however, show wide variability in their ingestion of oral contrast agent, resulting in under-distention and poor visualization of the small bowel. Small-bowel polyps, strictures, and adhesions are under-detected, especially in patients with Crohn disease [1]. Newer techniques include CT enterography [25], MR enterography [6], and capsule endoscopy [712]. CT enterography uses a high-osmolar, neutral-density oral contrast agent to both distend the bowel and allow more detailed visualization of the mucosal enhancement patterns. Yet like conventional MDCT of the abdomen, it requires a meticulous oral preparation protocol and ingestion of oral contrast agent. Children are especially unlikely to tolerate ingestion in a short period of the high volume of oral contrast agent needed. Capsule endoscopy provides direct visualization of the intestinal mucosa via multiple radio-transmitted images (average of 55,000 [13]) as it passes through the bowel. Although this method provides excellent visualization of the mucosa, its use might not be safe in young children [14].

First popularized in the 1970s, fluoroscopic enteroclysis (FE) was designed to provide a volume challenge to distend and better visualize the small bowel when compared to the barium SBFT [15]. Compared to SBFT and abdominal CT, this technique has been shown in adults to improve the detection and definition of small-bowel pathology, including intraluminal masses, mucosal diseases, stenosis/stricture, fistula, and partial obstruction [1, 16]. CT enteroclysis (CTE) combines the advantages of MDCT and enteroclysis. MR enteroclysis (MRE) might be a valuable tool to diagnose small-bowel disease, as well, especially in children because of its lack of radiation [1723]. MRE, however, has several logistical hurdles, including nasoduodenal intubation within the MRI room, motion artefacts (cardiovascular, respiratory, and bowel peristalsis), and breath-holding for sequences that last 20–30 s.

FE and CTE are diagnostic tools for the detection of small-bowel pathology, particularly strictures in children with Crohn disease. Several studies in the literature have focused on their utility and safety in adult populations, but no studies to our knowledge have described their use in children. We discuss here our fluoroscopic and CT techniques and sedation protocol in children. We also review the clinical history and indications for enteroclysis, document its safety and feasibility, and evaluate its added value by comparing the results of enteroclysis with those of other imaging modalities and with patient outcomes using clinical, surgical, and pathological follow-up.

Materials and methods

Population

This study was approved by our institutional review board with a waiver of informed consent. The retrospective study included three large teaching hospitals. From the radiology information system, we identified all FE and CTE studies performed on children, 18 years of age and younger, between January 2002 and March 2007. The electronic records of children who had received enteroclysis were reviewed, and pertinent findings from the clinical and surgical history prior to and following the enteroclysis were recorded, including the reported indications for enteroclysis, the findings of enteroclysis, history of known inflammatory bowel disease (IBD), and whether action had been taken based on the findings of the enteroclysis. Although we understand that using the reported indication might not reflect the most comprehensive reason as to why the study was performed, we thought it still added important comparative information. We reviewed the radiology nursing and clinical records of these patients for any complications from sedation or anesthesia.

Enteroclysis preparation

The enteroclysis procedure was described to the child and family prior to parental informed consent. All patients had nothing to eat or drink (NPO) after midnight the evening prior to the procedure. In order to minimize anxiety, children were allowed to bring their favorite stuffed toy animal with them and play music in the background. The radiology nurse provided the primary contact and information to the family for the procedure. All children were monitored by the nurse and with pulse oximetry and automatic blood pressure monitoring.

Sedation

A peripheral intravenous cannula was placed and 1 mg midazolam (Versed, Baxter Healthcare, Deerfield, IL; maximum dose 0.1 mg/kg), 25 µg fentanyl citrate (Fentanyl, Hospira, Lake Forest, IL; maximum dose 3 µg/kg), and metoclopramide were administered intravenously at the beginning of the procedure. Lidocaine 2% jelly was then placed in the nasal passage via syringe to ensure lubrication for the nasoenteric tube. Sedation was titrated to the child’s anxiety and discomfort throughout the procedure.

Rarely, the child did not meet the criteria for sedation by the radiology nursing staff and underwent either general anesthesia or did not undergo sedation.

Fluoroscopic enteroclysis technique

After initial sedation, a nasoduodenal catheter (MEC; Cook, Bloomington, IN) was inserted. For most children, a 13F tube was advanced through the pylorus to the fourth portion of the duodenum under fluoroscopic guidance. For small children younger than 8 years, a smaller balloon angiography catheter from the interventional radiology suite was used. The catheter balloon was initially inflated to 15–20 ml to ensure filling of the bowel lumen and prevent backflow of contrast agent into the stomach. Prior to infusion of contrast agent into the small bowel, intravenous glucagon was given. The enteroclysis contrast agent mixture used for FE varied by hospital and over time but included barium, methyl cellulose (until it was removed from the market), and iodinated contrast agents. The majority of patients, however, received contrast agent made of one-third 18% diatrizoate meglumine (Cystografin; Bracco, Princeton, NJ) and two-thirds 0.1% barium sulfate (VoLumen; E-Z-EM, Lake Success, NY). VoLumen has a radiographic density of about 20 HU. It contains sorbitol to maintain luminal distention and guar gum to increase viscosity [24].

The nasoenteric tube was connected to a hydraulic pump (RS-7800 Minipump; Renal Systems, Minneapolis, MN), and the enteral agent infused initially at a rate of 60 ml/min in order to assess contrast flow and bowel motility. If contrast agent refluxed into the stomach, the balloon was inflated further and the tube was repositioned. The contrast agent infusion rate was increased incrementally to a maximum of 120 ml/min (less in young children) under fluoroscopic observation [25]. If the child reported abdominal pain or the contrast agent advance slowed during infusion, the contrast agent infusion rate was decreased. The total infusion volume in older children was approximately 1,500 ml but varied widely depending on the child’s size, bowel distention, and history of bowel resection. During FE without CTE, the child was examined with multiple compression and noncompression spot fluoroscopic images during the infusion of contrast agent until contrast was seen in the cecum. The medical records and imaging reports were reviewed for complications including bowel perforation and bleeding.

CT enteroclysis technique

When the contrast agent reached the cecum, the infusion was stopped and the child was transported to the CT facility. The parameters for MDCT of the abdomen and pelvis were identical to the routine low-dose, weight-based parameters for conventional CT imaging. Images were acquired on a 4-, 16-, 40-, or 64-slice CT scanner (GE Medical Systems, Waukesha, WI, or Philips Medical Systems, Cleveland, OH). Imaging parameters included 120 kVp, 2.5- to 5-mm slice thickness with the current protocol using volumetric acquisition and reconstruction in all three planes. When the FE or CTE was complete, the balloon was deflated, and the stomach was aspirated. The child was then transferred to the radiology recovery area to be cared for by the radiology nursing or anesthesia staff.

Image analysis and comparison

Fluoroscopic and CT images were interpreted on PACS. CTE images were read using a window of 1,180 and level of 160 by a pediatric or gastrointestinal radiologist, depending on the study hospital location.

For this study, we reviewed all available imaging and diagnostic procedures within a 2-month period both preceding and following enteroclysis, including SBFT and abdominal CT images, and surgical and pathology results, in order to determine the final diagnosis and patient outcome. We reviewed pathology reports from surgery and endoscopic biopsies, and the reports were used for findings and diagnosis. We determined whether the imaging comparisons agreed, disagreed, or were noncomparable. Noncomparable findings were those that were not in the organ systems examined by both modalities. For example, an abdominal CT scan would evaluate all of the bowel as well as the solid organs, whereas the comparison FE imaging would be limited to the small bowel and did not evaluate the stomach, colon, or solid organs.

Data analysis

All data were tabulated in an Excel spreadsheet for reporting. We compared the indications for FE and CTE and their findings using chi-squared analysis. P values ≤0.05 were considered significant.

Results

Patients

We identified 113 FE and 74 CTE examinations in 176 children during the study period (Table 1). The demographics of children receiving FE and CTE were similar. Nine children had more than one examination. The children had a mean age of 14.2 years, a median age of 15 years, and an age range 3–18 years. There were 89 boys (51%). There were 48 children with Crohn disease and 6 with ulcerative colitis, although six children with Crohn disease had two FE examinations, and one child with Crohn disease had three FE examinations and one CTE examination during the study period.
Table 1

Demographic data for the children undergoing enteroclysis

 

FE studies (n = 113)

CTE studies (n = 74)

Age (years)

Mean

All patients

14

14.5

 Boys

13.2

14.7

 Girls

14.8

14.4

Range

8–18

3–18

Sex, n (%)

Boys

57 (50)

31 (42)

 Girls

56 (50)

43 (58)

Prior abdominal surgery, n (%)

16 (14)

15 (20)

Known Crohn disease, n (%)

35 (31)

20 (27)

Known ulcerative colitis, n (%)

1 (1)

5 (7)

The mean age of those undergoing FE was 14 years (median 15 years, range 8–18 years), and 57 (50%) were boys. Of those undergoing FE, 16 (14%) had prior abdominal surgery, 35 (31%) had known Crohn disease, and 1 (1%) had known ulcerative colitis. The mean age of those undergoing CTE was 14.5 years (median 15 years, range 3–18 years), and 32 (43%) were boys. Of those undergoing CTE, 15 (21%) had prior abdominal surgery, 20 (28%) had known Crohn disease, and five (7%) had known ulcerative colitis.

Feasibility and safety

Of the 113 FE studies, 112 were performed without difficulty and of the 73 CTE studies, 72 were performed without difficulty. One FE study was terminated because of patient anxiety, and one CTE study was terminated due to patient discomfort during the infusion of contrast agent into the small bowel. There were no complications.

Of 112 FE studies, 111 (99%) were performed under conscious sedation as per the pediatric enteroclysis protocol at the hospital and of 74 CTE studies, 67 (91%) were performed under conscious sedation. Four CTE studies were performed under general anesthesia because of a previous paradoxical response to Versed, young age (3 and 5 years), and severe autism with oppositional defiant disorder. One CTE study was performed without sedation or anesthesia because there was a mature tract to the stomach into which a Frederick-Miller tube was inserted. Using radiology nursing and medical chart reviews, we were unable to determine why one CTE study and one FE study were performed under general anesthesia and why one CTE study was performed without conscious sedation. There were no complications from either sedation or anesthesia.

Enteroclysis indications

The most common indications for FE and CTE (Table 2) were: known Crohn disease (FE 43 of 112, 38%; CTE 21 of 72, 29%), abdominal pain (FE 29 of 112, 26%; CTE 19 of 72, 26%), suspected small-bowel obstruction (FE 6 of 112, 5%; CTE 16 of 72, 22%), and suspected Crohn disease (FE 6 of 112, 5%; CTE 6 of 72, 8%). Gastrointestinal bleeding was the indication for 13 of 112 FE studies (12%) and 3 of 74 CTE (3%) studies. Known ulcerative colitis was an indication for 1 of 112 FE (1%) studies and 5 of 74 CTE (7%) studies. There were significantly more referrals for CTE than for FE for small-bowel obstruction (P = 0.001). Otherwise, the referral indications were similar.
Table 2

Clinical indications for FE (n = 112) and CTE (n = 74) studies

Indications

FE, n (%)

CTE, n (%)

Main

Crohn disease (known)

43 (38)

21 (28)

Abdominal pain

29 (26)

19 (26)

Gastrointestinal bleeding

13 (12)

3 (4)

Suspected small-bowel obstruction

6 (5)

16 (22)*

Suspected Crohn disease

6 (5)

6 (8)

Other

Chronic diarrhea

3 (3)

Weight loss

3 (3)

1 (1)

Irritable bowel syndrome

3 (3)

Colitis

2 (2)

Nausea/vomiting – no surgical history

1 (1)

1 (1)

Duodenitis

1 (1)

Familial polyposis syndrome

1 (1)

Juvenile polyposis syndrome

1 (1)

Protein losing enteropathy

1 (1)

Ulcerative colitis (known)

1 (1)

5 (7)

Iliac adenopathy and jejunal thickening

1 (1)

*P = 0.001

Fluoroscopic and CT enteroclysis findings

Of 112 FE studies, 60 (54%) were normal and of 74 CTE studies, 33 (45%) were normal (Fig. 1). While more CTE studies were abnormal than FE studies, the difference was not significant (P = 0.20). Active Crohn disease and its complications (FE 38 of 112, 34%; CTE 20 of 74, 27%) and partial small-bowel obstruction (FE 3 of 112, 3%; CTE 7 of 74, 10%) were the most common diagnoses for FE and CTE, respectively (Figs. 2, 3, 4, 5, 6; Table 3). Small-bowel obstruction was found in significantly more CTE studies than FE studies (P = 0.01). Other diagnoses by FE included adhesions without evidence of obstruction (2, 2%), known malrotation (1, 1%; Fig. 7), adenocarcinoma of the sigmoid colon and multiple colonic polyps (1, 1%; Fig. 8), ileal polyps (1, 1%), small-bowel/colon fistula (1, 1%), duplication cyst causing small-bowel intussusception (1, 1%; Fig. 9), and Meckel diverticulum (1, 1%; Fig. 10). Other diagnoses by CTE included ulcerative colitis (4, 6%), newly diagnosed malrotation with obstructing bands (2, 3%; Fig. 11), adhesions without and with evidence of obstruction (1, 1%; Figs. 12 and 13, respectively), left paraduodenal hernia (1, 1%; Fig. 14), and known malrotation (1, 1%).
https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig1_HTML.jpg
Fig. 1

Normal enteroclysis and CT enteroclysis examinations. a The small bowel is distended on the double contrast enteroclysis, providing exquisite detail of the normal mucosa. b CT enteroclysis. Coronal reconstruction MDCT image from a normal CTE study in an adolescent with suspected partial bowel obstruction. Note the small-bowel distention and mucosal detail produced by the large volume of positive contrast medium used for enteroclysis

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig2_HTML.jpg
Fig. 2

Partial obstruction, sinus tract formation and microabscesses in a 13-year-old girl with Crohn disease and a 2-week history of abdominal pain, fever and weight loss. Axial CT enteroclysis images. a There is a long segment stricture in the distal ileum with a moderately thickened and irregular wall (curved arrow) as well as intramural sinus tracts. b There is a small, ovoid fluid collection arising from the ileal wall representing a microabscess (open arrow). c Small linear areas of contrast within the wall indicate intramural sinus tract (open arrow). There is contrast medium filling of a normal appendix. Note the dilated small bowel proximal to the thickened ileum (curved arrow). The child subsequently underwent resection of the strictured distal ileum

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig3_HTML.jpg
Fig. 3

Sinus tract formation in a 15-year-old boy with active Crohn disease and abdominal pain. a Axial image from a CTE study demonstrates severe wall thickening and luminal narrowing of the distal ileum. b Coronal reconstruction image from a CTE study shows intramural sinus tracts involving this ileal bowel loop. The child subsequently underwent surgical resection

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig4_HTML.jpg
Fig. 4

Jejunoileal fistula in a 12-year-old girl with Crohn disease. a Spot image from an FE study shows early filling of an ileal loop (arrow) from dilated jejunal loops, confirming a fistula. b, c Axial CTE images show several small contrast-opacified fistulae (arrows) at the level of the umbilicus. Note the dilated small bowel proximal to the thick-walled, inflamed bowel. d Sagittal reconstruction image from a CTE study helps to confirm the findings and to localize the extent and location of the fistulae (arrow)

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig5_HTML.jpg
Fig. 5

Crohn disease with phlegmon and subsequent structuring in a 17-year-old girl with vomiting and recurrent abdominal pain. a Conventional axial MDCT image shows a heterogeneous air and fluid collection with surrounding mesenteric inflammatory changes that represent a phlegmon (arrow). b Coronal reconstruction MDCT image shows a small-bowel loop encased by the phlegmon. The girl was treated with antibiotics and her symptoms resolved. c Coronal reconstruction image from a CTE study performed weeks later for recurrent abdominal pain. The image demonstrates interval formation of a tight, short stricture (arrow) in the same segment of bowel. Note the smooth, thin-walled appearance of the stricture, as opposed to the low-grade partial small-bowel obstruction proximal to the stricture. This sequelae of Crohn disease illustrates the often rapid progression of disease. The girl underwent surgical resection of the stricture

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig6_HTML.jpg
Fig. 6

Crohn disease in a 12-year-old girl with new onset of vomiting and pain. a The initial enteroclysis image is normal. b A later enteroclysis image shows the terminal ileum is strictured (arrow). It produces dilation of the ileum proximal to it and required surgical resection. Two years later, she had recurrent symptoms of pain and fever and underwent a CT scan. c The right colon near the hepatic flexure is thick-walled and narrowed (arrow). d Near the mid-transverse colon, there is active inflammation of the colon and small, multiloculated abscesses (arrow). The strictured right colon was resected

Table 3

Primary diagnoses by FE (n = 112) and CTE (n = 72)

Diagnoses

FE, n (%)

CTE, n (%)

Main

Normal

60 (54)

33 (46)*

Crohn disease

38 (34)

20 (28)

Partial small-bowel obstruction

3 (3)

7 (10)**

Other

Lymphoid hyperplasia

5 (4)

1 (1)

Adhesions without obstruction

2 (2)

1 (1)

Polyps

2 (3)

Small-bowel polyps

1 (1)

Adenocarcinoma and polyps in colon

1 (1)

Wall thickening

Ileum

1 (1)

Jejunum

1 (1)

Meckel diverticulum

1 (1)

Malrotation

Known

1 (1)

1 (1)

Unknown

2 (3)

Small-bowel/colon fistula

1 (1)

Ulcerative colitis

4 (6)

Hernia

Left paraduodenal

1 (1)

Incisional

1 (1)

*P = 0.20, **P = 0.01

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig7_HTML.jpg
Fig. 7

Malrotation in a 3-year-old girl with heterotaxy syndrome (polysplenia). a Scout image from the CTE study shows a large number of small-bowel loops in the right hemiabdomen and stool in the colon located in the left abdomen. b Coronal reconstruction image from a CTE study demonstrates small bowel in the right hemiabdomen with early filling of the cecum in the left lower quadrant (large arrow). There is normal reflux of contrast medium into the biliary tree from a previous Kasai procedure for biliary atresia (small arrow)

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig8_HTML.jpg
Fig. 8

Multiple colonic polyps and sigmoid adenocarcinoma in a teenage girl with recent hematochezia. Attempted colonoscopy revealed a mass at the sigmoid colon that could not be passed. The initial biopsy showed adenomatous polyp. a, b Coronal reconstructed images from a CTE study show an “apple core” lesion in the sigmoid colon (open arrows) that was resected. Pathology showed adenocarcinoma. There were multiple smaller solid polyps throughout the colon (closed arrows) and rectum

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig9_HTML.jpg
Fig. 9

Enteroclysis in a 10-year-old boy with anemia and abdominal pain. The image shows a duplication cyst with small-bowel intussusception (arrow). These findings were confirmed at surgery

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig10_HTML.jpg
Fig. 10

A 17-year-old boy who presented with painless rectal bleeding. He had a negative colonoscopy, Meckel scan and tagged RBC scan. Spot fluoroscopic image from a double-contrast enteroclysis shows a focal outpouching from a posterior pelvic ileal loop (arrow) consistent with a Meckel diverticulum. At surgery, the outpouching was confirmed to be a Meckel diverticulum

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig11_HTML.jpg
Fig. 11

CTE images of longstanding partial small-bowel obstruction from malrotation with Ladd bands in an 8-year-old boy who presented with chronic abdominal pain, chronic anemia and diarrhea. He was clinically suspected to have Crohn disease. a Axial CTE image demonstrates marked dilation of the distal small bowel (solid arrow) with areas of nondistended small bowel (open arrow). b, c Axial CTE images demonstrate markedly dilated and distorted, featureless segments of small bowel with adhesive bands, most pronounced in the right lower quadrant (open arrow). d Coronal CTE image shows the unopacified cecum in the right upper quadrant (curved arrow). The small bowel is distorted and distended with tethering and narrowing from Ladd bands (straight arrows). The boy underwent a Ladd procedure and resection of the right lower quadrant abnormal small bowel

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig12_HTML.jpg
Fig. 12

Enteroparietal adhesions in a 13-year-old girl with a history of left nephrectomy and radiation for Wilms tumor. a Axial CTE image just below the umbilicus shows abnormal approximation of multiple bowel loops to the anterior parietal peritoneum consistent with adhesive disease (arrows). Normally, a thin fat plane should be seen between bowel and adjacent parietal peritoneum. b Sagittal reconstruction image from a CTE study again demonstrates abnormal enteroparietal adhesions near the umbilicus (arrow). There was no dilated bowel to suggest obstruction and the child did not undergo surgery

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig13_HTML.jpg
Fig. 13

Adhesions causing high-grade partial small-bowel obstruction in a 7-year-old girl who had had an appendectomy 5 months previously. a, b Axial CTE images show dilated small-bowel loops with abrupt caliber change (a, copen arrows) in the right lower quadrant with normal caliber distal ileal loops (bcurved arrow). c, d Coronal reconstruction images from a CTE study show a tethered appearance of obstructed small-bowel loops from adhesions (solid arrows) with small-bowel caliber change from dilated to collapsed bowel distally (open arrow). Adhesions were resected at surgery

https://static-content.springer.com/image/art%3A10.1007%2Fs00247-008-0754-8/MediaObjects/247_2008_754_Fig14_HTML.jpg
Fig. 14

Left paraduodenal hernia in a 10-year-old girl with chronic intermittent abdominal pain and vomiting. a, b Coronal and axial CT images show normal small bowel and the study was normal. c Supine AP spot image from the fluoroscopic phase of a CTE study shows multiple dilated jejunal loops contained by a sac in the left upper quadrant (arrows) consistent with a left paraduodenal hernia. Note the nondilated small bowel distal to the hernia sac. The findings were confirmed at surgery

With regard to diagnoses unrelated to the small bowel, FE studies showed one child with possible colonic obstruction and one child with appendicitis. CTE studies showed three children (4%) with hiatal hernia and two children (3%) with ovarian cysts, as well as one child each with renal cysts, polysplenia, splenic infarcts, diffuse osteosclerosis, left renal duplication, colonic pneumatosis, bilateral pars defects, and pancolitis.

Crohn disease was found in 38 FE studies in 31 children, with 6 (16%) being new diagnoses. Of the children with established diagnoses of Crohn disease, 9 of 25 (36%) had small-bowel stenosis or partial small-bowel obstructions, one had a sinus tract, and one had a fistula. Two patients had multiple abnormalities. A CTE study was carried out in 20 children with Crohn disease; none was a new diagnosis. Of the 20 children, 12 (60%) had small-bowel stenosis or partial small-bowel obstructions, 4 (20%) had sinus tracts, 3 (15%) had fistulas, and 2 (10%) had phlegmons surrounding active small-bowel disease. Eight children undergoing a CTE study had multiple abnormalities (Table 4).
Table 4

Crohn disease findings on FE (n = 38) and CTE (n = 20) studies

Crohn disease diagnoses

FE, n (%)

CTE, n (%)

New diagnosis

6 (16)

0

Stricture / partial SBO

11 (29)

12 (60)

Sinus tract

1 (3)

4 (20)

Fistula

1 (3)

3 (15)

Phlegmon

0

2 (10)

Number with multiple findings

2 (5)

8 (40)

Ulcerative colitis was diagnosed in four children by CTE. Two of the four children had partial small-bowel obstructions, one had small bowel to anterior abdominal wall adhesions without evidence of obstruction, and one had a thickened colonic wall. None of the four children had multiple abnormalities. Ulcerative colitis was not diagnosed by FE.

Abdominal CT correlation

Of 112 children, 21 (19%) had conventional abdominal/pelvic CT scans performed within 2 months of FE. The FE findings were concordant with the CT findings in 17 (81%), discordant in 3 (14%), and noncomparable in 1 (5%). In the child with noncomparable findings, CT showed proctocolitis and FE showed a normal small bowel. Of the three children with discordant findings, in one child CT showed small-bowel inflammation from Crohn disease, whereas FE showed a stricture at this site. In one child CT showed ascites but no active Crohn disease, whereas FE showed active Crohn disease of the small bowel. In the third child CT showed multiple sigmoid diverticuli and no active Crohn disease, whereas FE showed active Crohn disease of the terminal ileum without evidence of fistula (Table 5). These three discordant cases demonstrate that FE provided additional information to CT.
Table 5

Imaging correlations within 2 months of enteroclysis

Comparator modality

No. of studies compared

Enteroclysis

Correlation, n (%)

Concordant

Discordant

Noncomparable

Abdominal CT

21

FE

17 (81)

3 (14)

1 (5)

21

CTE

16 (76)

3 (14)

2 (10)

SBFT

4

FE

2 (50)

1 (25)

1 (25)

11

CTE

9 (82)

2 (9)

0

Surgical/pathological

67

FE

39 (58)

1 (1)

28 (42)

34

CTE

27 (79)

1 (3)

6 (18)

Of 74 children, 21 (28%) had conventional CT scans within 2 months of CTE. CTE findings were concordant with CT findings in 16 (76%), discordant in 3 (14%), and noncomparable in 2 (10%). The noncomparable studies showed interval resolution of lymphadenopathy over 18 days and mild ascites over 2 days between the CT scan and CTE performed subsequently. No treatment was administered to either patient during these time periods. In the three children with discordant studies, CT showed no abnormality and the subsequent CTE showed abnormalities (one child each with jejunal wall thickening, Crohn disease of the mid-ileum, and partial small-bowel obstruction; Table 5). These three discordant results show that CTE provided additional information to CT.

SBFT correlation

SBFT was performed in four children within 2 months of FE. The findings were concordant in two children (50%), discordant in one (25%), and noncomparable in one (25%). In the child with noncomparable findings, SBFT showed gastric reflux and FE showed no abnormality. In the child with discordant findings, SBFT showed Crohn disease of the distal ileum with cobblestoning and stricture and FE showed evidence of stricture 2 months later. The diagnosis of Crohn disease had been previously confirmed by biopsy. It is possible that the distal ileal disease resolved in the interval with therapy (Table 5).

SBFT was performed in 11 children within 2 months of CTE. The findings were concordant in nine children (82%) and discordant in two. In the child with discordant findings, SBFT showed nonobstructive narrowing of the ileum and CTE showed no abnormality. The narrowing was later thought to represent focally collapsed bowel or spasm, as it was not present on any previous or later studies. In the other child with discordant findings, SBFT showed a cecal fistula and CTE showed ulcerative colitis with colonic wall-thickening but no fistula. The child underwent a total colectomy 27 days after CTE and 43 days after SBFT for ulcerative colitis. Pathology did not show evidence of a cecal fistula (Table 5).

Surgical/pathological correlation

Of 112 surgical or pathological procedures (from endoscopic biopsies), 67 (59%) were performed within 2 months of FE. Endoscopy or colonoscopy was performed in 17 children. The surgical/pathological results were concordant in 39 children (58%), discordant in 1, and noncomparable in 28 (42%) as the surgical/pathological specimens were not small-bowel samples. In the child with discordant results, FE showed no abnormality, but a terminal ileal stricture was found and resected at surgery. Although not considered discordant, we note that in one child FE showed a polyp that was found to be an ileal duplication cyst at surgery (Table 5).

Of 74 surgical/pathological procedures, 24 (46%) were performed within 2 months of CTE. The surgical/pathological results were concordant in 27 children (79%), discordant in 1, and noncomparable in 6 (18%) as the surgical/pathological specimens were not small-bowel samples. In the child with discordant results, CTE showed no abnormality but subsequent ileal biopsy performed during colonoscopy on the same day was consistent with Crohn disease (Table 5).

Overall, 14 of 112 children who underwent FE and 21 of 74 children who underwent CTE also underwent previously unscheduled surgery within 2 months of the FE or CTE because of the findings of these examinations. In particular, segmental small-bowel resections were performed in nine, exploratory adhesolysis in four, and colectomy in four.

Discussion

Enteroclysis was found to be safe and feasible in a large and diverse pediatric population. New diagnoses were made in 20 of 112 FE studies (19%) and 18 of 74 CTE studies (24%). Enteroclysis was mainly utilized to evaluate patients with known or suspected Crohn disease and partial small-bowel obstruction. The CTE results sometimes provided a definitive answer about whether the small bowel, when volume-challenged, had a partial obstruction from internal hernia, adhesions or stricture. Compared to SBFT and conventional CT, FE and CTE added diagnostic information. This was particularly helpful in children with active IBD of the small bowel, such as strictures, sinus tracts, fistulas, and phlegmon. CTE provided additional information about both small-bowel and extraintestinal pathology to FE and altered the surgical management of 28% of patients, as compared with 12% of patients who underwent FE.

Although this study focused on the pathological findings of enteroclysis in children, the importance of a normal study in excluding a partial small-bowel obstruction or new onset of Crohn disease is easily overlooked. Gastroenterologists use the enteroclysis study to rule out pathology, particularly partial small-bowel obstruction, in patients with chronic abdominal pain. One child with Crohn disease had abdominal pain out of proportion to her clinical examination and laboratory findings. We reassured the pediatric gastroenterologist that a stricture or partial bowel obstruction was not causing her symptoms. For some other children, the pediatric gastroenterologist, surgeon, and radiologist must work together to build evidence about the need for therapeutic intervention. For 3 years Barloon et al. [26] followed 83 patients who had a normal enteroclysis to assess its negative predictive value. Only six were found to have small-bowel pathology, meaning that the enteroclysis had 92% specificity for ruling in any disease [26].

Crohn disease is a common cause of small-bowel pathology in both adults and children [27], yet there are no optimal disease activity markers. As such, its evaluation with imaging is critical. Capsule endoscopy is equivalent or superior to other modalities in the evaluation of known ileal Crohn disease [712]. It is the method of choice in patients with unexplained gastrointestinal bleeding or a hereditary polyposis syndrome [11, 28]. However, capsule endoscopy cannot be used when there is concern for stricture or partial bowel obstruction, or to evaluate extraintestinal complications. While capsule endoscopy is increasingly used in adults, it cannot be used in young children because of the smaller size of the bowel [12, 14].

Alternative imaging methods for evaluation of the small bowel are less invasive than enteroclysis. CT enterography is a less invasive method than enteroclysis that depicts inflammation of the small bowel associated with Crohn disease and its findings correlate more closely with the ileoscopic and histological severity in Crohn disease than those of SBFT [25]. MR enterography avoids the radiation exposure of CT, but has more issues with motion artefacts [6, 29]. Similarly, MR enteroclysis is being developed to avoid ionizing radiation, but it shows more bowel, vascular, and respiratory motion artefacts [30]. In our experience, the main limitation of enterography in children is the lack of adequate distention of the small bowel, especially the distal ileum, often the area of interest in patients with Crohn disease. We have found that children are usually unwilling to drink large amounts of oral contrast agent over the short time interval recommended for optimal imaging.

In order to identify low-grade obstructions, small-bowel volume challenge is crucial to show small-caliber changes [31]. In adults, CTE is the imaging modality of choice for the localization and characterization of small-bowel neoplasms [8, 32, 33]. Enteroclysis in adults might also be the most accurate imaging method for the diagnosis of complications from Crohn disease and small-bowel obstructions [1, 21]. Its disadvantage is that a nasoenteric tube is invasive and its use requires sedation. An added disadvantage, especially because of the increased risk in children, is radiation exposure, principally from the CT component. Radiation doses related to enteroclysis and CTE are as significant an issue as they are in the use of abdominal and pelvic CT studies in these patients. It is important to minimize fluoroscopy time and perform FE and CTE using the principle of ALARA [34]. We use a low-dose technique that is weight-based, identical to that used in our conventional abdominal MDCT. Because many patients with Crohn disease undergo repeated CT imaging for evaluation of symptoms and signs, performing CTE at the same dose is no less justified and provides more clinical information in selected patients. Although the technique for FE varies, one study has demonstrated that the radiation dose is lower for CTE than for FE in adults [35]. In either case, performing serial examinations that expose children to ionizing radiation should be quantified so that cumulative radiation exposure can be determined and adequate patient safety maintained. Currently, we do not have a system in place to monitor and tabulate total radiation exposure to individual patients for any of our ionizing radiation examinations.

The main advantage of CTE over other imaging modalities is depiction of luminal, mural, and extramural abnormalities. With improved MDCT technology and the advent of CTE in 1994, thinner collimation and volumetric image acquisition allows review in all three planes [17, 3639]. CTE has been found to be superior to FE in the diagnosis and characterization of mucosal abnormalities, intramural and extramural Crohn disease, and low-grade small-bowel obstruction [17, 19]. Because of these findings and our own experience with both FE and CTE, we have performed CTE exclusively in all children since November 2005.

In complex clinical situations, the pediatric gastroenterologist, pediatric surgeon, and pediatric radiologist must make decisions about the appropriate treatment for children with abdominal pain. CTE is an important tool in the evaluation of the small bowel in children. The radiologist, the radiology technologist, and the sedation nurse work as a team to complete the procedure safely, allay patient and family anxiety, and discuss the enteroclysis results at the end of the study.

Conclusion

Both FE and CTE were well tolerated and safe and provided additional information to that provided by SBFT and abdominal CT in a number of children. The pediatric gastroenterologist and pediatric surgeon use the results of FE and CTE for treatment planning. Compared to FE, CTE showed more abnormalities both in the small bowel and outside the bowel. For this reason, we perform CTE rather than FE. We believe that CTE has an important role in evaluating the small bowel in children, particularly in those with Crohn disease.

Copyright information

© Springer-Verlag 2008