Reilly, Norelle R.*; Aguilar, Kathleen M.†; Green, Peter H.†
Intussusception (IS) is the second most common cause of gastrointestinal obstruction in young children, and the most common cause of small bowel obstruction in children ages 3 months to 5 years with an approximate incidence of 22 to 56 cases per 100,000 per year (1–3). Occasionally, a lead point such as Meckel diverticulum or lymphoma may be found, but in 90% to 95% of pediatric cases, no cause can be identified (1,4). Despite the fact that IS is relatively rare beyond childhood (5–7), an association with celiac disease (CD) has been described in adults (8–11). In contrast, descriptions of ISs in children suspected to have been caused by underlying CD are limited to case reports (12–16). There are, however, no studies determining how frequently ISs occur among children with CD. Our aim was, therefore, to determine whether the frequency of ISs in children with known CD is greater than that among our pediatric population.
We reviewed the records of 318 patients contained in a registry of children with biopsy-proven CD seen at our center from 2000 to 2010. These records were reviewed for abdominal imaging studies (ultrasound, magnetic resonance imaging, computed tomography [CT], or small bowel series) and we noted those patients whose imaging revealed an IS. Using a clinical data repository maintained by our hospital, we also identified children ages 6 months to 18 years who were diagnosed with IS at our center during the same period. A cutoff of 6 months of age was used because of the typical start of grains in the infant diet at this age in the United States. We reviewed these patients’ records for a diagnosis of CD preceding or following the IS, as well as for histologic or serologic evidence of possible CD (increased liver function tests; iron-deficiency anemia; hypoalbuminemia; and/or positive tissue transglutaminase IgA, anti-endomysial IgA, or anti-gliadin IgA/IgG). These records were further reviewed for recurrent presentation to the emergency department or pediatrician for abdominal pain.
Exact 95% binomial confidence intervals were calculated for each IS prevalence value. The present study was approved by the institutional review board of Columbia University Medical Center.
Patients With IS
For the 303,612 pediatric patients (mean age 7 ± 6.2 years) seen at our center from January 1, 2000 to January 1, 2010, 23,692 abdominal radiographic studies (CT, sonogram, magnetic resonance imaging, barium enema, small bowel series) were performed for children ages 6 months to 18 years, identifying 226 cases of IS in 216 patients. This translated into ISs diagnosed in 0.07% of children seen at our center during this time frame (95% CI 0.06%–0.08%). In 210 cases (200 patients), the IS was idiopathic. In the remainder of cases, a potential underlying cause or complicating factor was identified (Meckel's diverticulum , rotavirus infection , inflammatory bowel disease , Henoch-Schönlein purpura , Peutz-Jegher syndrome , jejunostomy/gastrojejunostomy , nasoduodenal tube , history of prior abdominal surgery ).
Thirty patients diagnosed with idiopathic IS returned to the emergency department or primary care office for evaluation of abdominal pain following their initial IS. In 10 cases of returning patients (33%), a second IS was identified and treated. Three patients without confirmed IS recurrence returned 2 or more times with abdominal pain. When we reviewed the laboratory tests of patients experiencing IS, we identified 11 patients with iron-deficiency anemia, 2 of which had return visits for abdominal pain.
Testing for CD with serology or esophagogastroduodenoscopy was performed for 4 of 200 patients with idiopathic IS: 1 patient with pain recurrence, and 3 patients with no apparent additional risk factors. All of the 4 patients’ celiac tests were negative. None of the patients with CD and ISs were tested, nor were patients with a history of IS and additional findings of iron deficiency anemia.
Patients With Known CD
Of the 318 patients with biopsy-proven CD in our registry, 53 had undergone a total of 77 abdominal imaging studies (Table 1) (1–3,5–9,17). The majority of those imaged (66%) were girls. Abdominal pain was the indication for 75% of studies; nearly half of these studies (48%) were performed before a diagnosis of CD was established. All of the episodes of IS were identified by ultrasound. IS represented the most frequent abnormal sonographic diagnosis among patients in the cohort (12.7%).
Four patients of 254 with complete records available (mean age 8.6 ± 4.9 years) had a history of IS (6 episodes total) (Table 2). All of the 4 patients experienced an IS before being treated for CD. In 3 patients, the IS prompted workup for CD and preceded the diagnosis by 9 months or less. The fourth patient was diagnosed with IS at age 5 months and was excluded from data analyses because it could not be ascertained whether he had been introduced to gluten at the time of his IS. Considering the remaining 3 patients, 1.2% of our cohort of children with CD experienced a known IS (95% CI 0.2–3.4).
Only 1 patient had laboratory abnormalities aside from positive celiac serologies (hypoalbuminemia and iron-deficiency anemia); the other 2 had no laboratory evidence of nutritional deficits. Normal body mass index (BMI) z scores were noted for patients with IS and available data. No patients with CD were diagnosed with an IS following treatment with a gluten-free diet.
Approximately 1200 to 1400 cases of IS occur annually in children in the United States alone (18). Although the majority of ISs in children are believed to be idiopathic, those with known etiologies are caused by lesions such as tumors or Meckel diverticula as well as associated conditions such as Henoch-Schönlein purpura (19), viral infections (20), and receipt of past versions of the oral rotavirus vaccine (21). There may be an increased risk among patients with Crohn disease, in whom case reports of ISs have been described (22), whereas in 1971, Holsclaw et al (23) described a 1% prevalence of IS among approximately 2200 patients with cystic fibrosis.
In the present study, we report that >1% of children with CD experienced a known IS before treatment with a gluten-free diet. This was similar to recent documentation of the prevalence among adults with CD (1.6%) (10). Given that not all of the children with CD and abdominal pain underwent abdominal imaging, this estimate refers only to cases clinically significant enough to warrant medical attention. The true prevalence of IS in this population may exceed this figure if one assumes that many patients with comparatively mild abdominal symptoms may have had transient, undiagnosed ISs. ISs among patients in our cohort of children with CD exceeded recent reports of the overall prevalence in American children (up to 0.037%) (3), as well as at our institution (approximately 0.07%).
An association between transient, nonobstructive ISs and CD has been described in the adult literature for >30 years, (24) some studies citing the prevalence of transient ISs among adults with CD and malabsorptive symptoms at 20% (8). One case report documents an IS in a women who underwent laparotomy because of a presumed tumor, who was later diagnosed as having mesenteric lymphadenopathy and underlying CD (8). Among adults, IS may also occur in the setting of type II refractory CD and is less common in uncomplicated CD (25). Transient small bowel ISs have been noted in children (26,27), although the association between these and CD is not well described to date in the pediatric literature.
The reason for the lack of pediatric data on this subject may be attributed to the fact that IS is a rare finding among adults, so an etiology may be sought more vigorously. In their recent 25-year review of adult patients diagnosed with IS, Onkendi et al (28) studied 196 patients with IS, 60% of which underwent CT scanning, and 61% of which required surgical management. Malignancy was found to be the cause in 22% of cases; CD accounted for 4%. In contrast, ISs are comparatively common in children, and usually considered to be idiopathic (1,4). Pediatricians may not feel compelled to probe further into predisposing factors of a child who otherwise appears healthy.
Data regarding an association between ISs and childhood CD have been limited to case reports and cite transient small bowel-small bowel ISs not requiring intervention (12–14,16). In our patient series, 1 child manifested multiple small bowel-small bowel and ileocolic ISs, whereas the other lesions were purely ileocolic. Furthermore, all of the cases of IS occurred in children with untreated CD, and in 3 of 4 cases IS was a presenting complaint prompting further evaluation. With the exception of 1 patient diagnosed as having CD at age 11 years whose IS occurred in infancy, all of the others were diagnosed as having CD weeks to months after the IS. The close temporal relationship of most cases with the patients’ diagnoses with CD suggests more than a coincidental association. In addition, the absence of IS among children treated for CD in our cohort should be noted; similarly, 64% of adult patients with CD and IS in a recent study were shown to have been diagnosed with an IS soon before or after diagnosis with CD (10).
Among patients diagnosed with IS at our institution, only 2% were tested for CD, including 1 patient with recurrent episodes. Our data have demonstrated that evidence of nutritional deficiencies, such as iron-deficiency anemia and low albumin, need not be present for a child with IS and CD. In addition, all of the patients with IS who were later diagnosed as having CD demonstrated normal growth patterns. Although red flags such as recurrent small bowel-small bowel ISs, failure to thrive, and nutritional deficiencies may facilitate a diagnosis of CD, an index of suspicion for CD should be maintained in the setting of IS despite absence of these criteria.
Whether transient ISs underlie some complaints of chronic or recurrent abdominal pain in children with CD is not known, nor is it known how often ISs underlie pediatric complaints of recurrent abdominal pain in general. Abdominal pain is a frequent presenting complaint in children diagnosed as having CD (25,27,29–31), although the etiology of the pain is unclear. Three of the 4 patients with IS had complained of chronic abdominal pain, which resolved following dietary treatment. Additionally, abdominal pain was the indication for 75% of abdominal imaging studies performed; however, the majority of patients in our cohort who had abdominal pain did not undergo radiologic imaging (62.5%). Prospective studies are warranted to investigate the etiology of abdominal pain in children with untreated CD, as well as in patients compliant with dietary therapy.
Limitations of the present study include its retrospective nature. A prospective design may better approximate the rate of ISs among children with CD, perhaps capturing some that are transient. Additionally, complete radiology records were not available for all of patients in our cohort. Finally, although the frequency of ISs at our institution was close to the prevalence in the United States, because ours is a quarternary care medical center, the experiences of our patients may differ from those of patients elsewhere in the United States or in other countries.
In conclusion, we describe a greater frequency of IS among children with CD than in a general pediatric population. CD may be an underlying cause of IS and should be considered even in well-appearing children, although particularly if nutritional deficiencies or growth failure are also seen or in the setting of baseline abdominal complaints. IS may be a cause of recurrent abdominal pain in children with CD; however, it does not recur after diagnosis and treatment of CD with a gluten-free diet. Prospective studies are required to further explore the relation of abdominal pain, IS, and CD.
The authors acknowledge Alla Babina of the Department of Biomedical Informatics at Columbia University for efforts in data collection.
1. Stringer MD, Pablot SM, Brereton RJ. Paediatric intussusception. Br J Surg
2. DiFiore JW. Intussusception. Semin Pediatr Surg
3. Parashar UD, Holman RC, Cummings KC, et al. Trends in intussusception-associated hospitalizations and deaths among US infants. Pediatrics
4. Lehnert T, Sorge I, Till H, et al. Intussusception in children—clinical presentation, diagnosis and management. Int J Colorectal Dis
5. Donhauser JL, Kelly EC. Intussusception in the adult. Am J Surg
6. Cotlar AM, Cohn I Jr. Intussusception in adults. Am J Surg
7. Agha FP. Intussusception in adults. AJR Am J Roentgenol
8. Cohen MD, Lintott DJ. Transient small bowel intussusception in adult coeliac disease. Clin Radiol
9. Dodds BF, Aguancha SI, Santamarina RM, et al. [Celiac disease presenting as an intestinal intussusception. Report of one case]. Rev Med Chil
10. Gonda TA, Khan SU, Cheng J, et al. Association of intussusception and celiac disease in adults. Dig Dis Sci
11. Tomei E, Diacinti D, Marini M, et al. Abdominal CT findings may suggest coeliac disease. Dig Liver Dis
12. Teitelbaum JE. Clinical quiz. Small bowel-small bowel intussusception. J Pediatr Gastroenterol Nutr
13. Germann R, Kuch M, Prinz K, et al. Celiac disease: an uncommon cause of recurrent intussusception. J Pediatr Gastroenterol Nutr
14. Mushtaq N, Marven S, Walker J, et al. Small bowel intussusception in celiac disease. J Pediatr Surg
15. Al Furaikh S, Al Zaben AA. Recurrent small bowel intussusceptions: an uncommon presentation of celiac disease in an Arab child. Trop Gastroenterol
16. Fishman DS, Chumpitazi BP, Ngo PD, et al. Small bowel intussusception in celiac disease: revisiting a classic association. J Pediatr Gastroenterol Nutr
17. Rashid M, Cranney A, Zarkadas M, et al. Celiac disease: evaluation of the diagnosis and dietary compliance in Canadian children. Pediatrics
18. Tate JE, Simonsen L, Viboud C, et al. Trends in intussusception hospitalizations among US infants, 1993–2004: implications for monitoring the safety of the new rotavirus vaccination program. Pediatrics
19. Ebert EC. Gastrointestinal manifestations of Henoch-Schonlein purpura. Dig Dis Sci
20. Lappalainen S, Ylitalo S, Arola A, et al. Simultaneous presence of human herpesvirus 6 and adenovirus infections in intestinal intussusception of young children. Acta Paediatr
21. Murphy TV, Gargiullo PM, Massoudi MS, et al. Intussusception among infants given an oral rotavirus vaccine. N Engl J Med
22. Knowles MC, Fishman EK, Kuhlman JE, et al. Transient intussusception in Crohn disease: CT evaluation. Radiology
1989; 170 (3 Pt 1):814.
23. Holsclaw DS, Rocmans C, Shwachman H. Intussusception in patients with cystic fibrosis. Pediatrics
24. Ruoff M, Lindner AE, Marshak RH. Intussusception in sprue. Am J Roentgenol Radium Ther Nucl Med
25. Lurz E, Scheidegger U, Spalinger J, et al. Clinical presentation of celiac disease and the diagnostic accuracy of serologic markers in children. Eur J Pediatr
26. Mateen MA, Saleem S, Rao PC, et al. Transient small bowel intussusceptions: ultrasound findings and clinical significance. Abdom Imaging
27. Kim JH. US features of transient small bowel intussusception in pediatric patients. Korean J Radiol
28. Onkendi EO, Grotz TE, Murray JA, et al. Adult intussusception in the last 25 years of modern imaging: is surgery still indicated? J Gastrointest Surg
29. McGowan KE, Castiglione DA, Butzner JD. The changing face of childhood celiac disease in North America: impact of serological testing. Pediatrics
30. Reilly NR, Aguilar K, Hassid BG, et al. Celiac disease in children with normal weight and overweight: clinical features and growth outcomes following a gluten-free diet. J Pediatr Gastroenterol Nutr
31. Westerbeek E, Mouat S, Wesley A, et al. Coeliac disease diagnosed at Starship Children's Hospital: 1999–2002. N Z Med J