Article, Pediatrics

Intussusception in traditional pediatric, nontraditional pediatric, and adult patients

Original Contribution

Intussusception in traditional pediatric, nontraditional pediatric, and adult patients?,??

Alexis A. Cochran MD, George L. Higgins III MD, Tania D. Strout RN, BSN, MS?

Maine Medical Center, Department of Emergency Medicine, Portland, ME 04102, USA

Received 4 November 2009; accepted 29 November 2009

Abstract

Study objectives: We sought to determine the rate of intussusception in 3 age groups (traditional pediatric-age [T], nontraditional pediatric-age [N], and adult-age [A]) and to compare group characteristics. Methods: We conducted a retrospective records review for patients discharged with diagnosis of intussusception between October 1999 and June 2008.

Results: Ninety-five cases of intussusception were diagnosed as follows: 61 T (64%), 12 N (13%), and 22 A (23%). Bloody stool was more common in T patients (P = .016). Air contrast enema (36%) and ultrasound (33%) were the most common diagnostic tests in T, whereas computed tomography was most common in N (83%) and A (68%) patients. bowel resection occurred more often in older (T) patients (P =

.001). The most frequent causative Pathologic conditions were adenitis (T), Peutz-Jeghers polyp (N), and carcinoma (A) and prior gastric bypass in 10 A patients.

Conclusions: The incidence of intussusception is substantially higher in nontraditional age groups than previously reported. Symptoms, management strategies, and causative pathologic conditions varied with age. All adults with intussusception require definitive diagnostic testing to determine the cause, given the concerning list of possibilities we observed.

(C) 2011

Introduction

Background

Intussusception is a gastrointestinal condition that involves the involution of a proximal portion of the bowel into a more distal portion, leading to inflammation and bowel obstruction. This is not an uncommon cause of abdominal pain and obstruction in young children, and there is general

? This work was presented at the 2009 American College of Emergency Physicians Research Forum, October 5, 2009.

?? This study was not externally funded.

* Corresponding author. Tel.: +1 207 662 7049; fax: +1 207 662 7025.

E-mail addresses: [email protected] (A.A. Cochran), [email protected] (G.L. Higgins), [email protected] (T.D. Strout).

familiarity with the presentation of intussusception in the pediatric population. Classic signs include episodic abdom- inal pain, bilious emesis, and “currant jelly” stools. Primary idiopathic intussusception accounts for 90% of Pediatric cases, with most intussusceptions being ileocolic in location. Most pediatric, Family Medicine, surgery, and emergency medicine texts site the typical age range of presentation for intussusception as being 2 months to 6 years old, with the peak incidence occurring between 5 and 12 months of age. These same texts frequently quote the occurrence of intussusception in patients outside this age range as “infrequent” or “rare.” For example, the 2007 edition of the Nelson Textbook of Pediatrics identifies intussusception as the “most common cause of intestinal obstruction between 3 months and 6 years of age,” whereas the 2010 edition of Rosen’s Emergency Medicine Concepts and Clinical

0735-6757/$ – see front matter (C) 2011 doi:10.1016/j.ajem.2009.11.023

Practice states that “only 5% of all intussusceptions occur in adults” [1,2].

Given this, it is reasonable to assume that making the diagnosis of intussusception in patients who lie outside the “traditional” pediatric age group may be more challenging for clinicians. Our anecdotal observation has supported this assumption. We have also observed that older children and adults who present to our emergency department (ED) with acute, undiagnosed abdominal pain and who ultimately prove to have intussusception as a cause of their symptoms are not as uncommon, when compared to young children, as general medical tenants would suggest. We hypothesized that symptomatic intussusception occurs more frequently in older pediatric and adult patients than has previously been appreciated and described. The goals of this investigation were to (a) determine the rate of intussusception in 3 age groups: traditional pediatric-age, nontraditional pediatric- age, and adult-age and (b) to compare symptoms, diagnostic and management strategies, and causative pathologic condi- tions between the 3 groups.

Materials and methods

Study design

We conducted a single institution, multiyear (October 1999-June 2008) retrospective health records survey for patients presenting to the ED who received a final imaging or surgically proven principle diagnosis of intussusception. The institutional review board at Maine Medical Center exempted the study and waived the requirement for written informed consent.

Setting

Maine Medical Center is a 600-bed tertiary care academic medical center that services a population of approximately 250 000 and provides care to approximately 59 000 ED patients and 32 000 in-patients annually.

Selection of subjects

The initial patient cohort was identified through a search of specific intussusception-related International Classifica- tion of Diseases and Related Health Problems, Ninth Edition, codes. Each patient record was then reviewed to ensure that acute intussusception was in fact the cause of the patient’s presenting complaint. Health records for patients with nonintussusception visits were excluded from the study.

Data collection

The following data elements were collected, as available, on all study subjects: age; sex; race; prior primary care

provider visits for same condition; prior ED visits for same condition; chief complaint; admission systolic blood pres- sure, diastolic blood pressure, pulse rate, and temperature; presence of vomiting; presence of diarrhea; presence of constipation; presence of bloody stools; admitting diagnosis; Pain severity; pain characteristic; pain duration; pain location; history of abdominal pain; prior time frame of pain pattern; prior frequency of pain pattern; prior known gastrointestinal condition; known family history of gastro- intestinal condition; admission white cell blood count, absolute Neutrophil count, hemoglobin level, hematocrit, glucose, blood urea nitrogen, creatinine level, and electro- lytes; diagnostic test; final therapeutic intervention; patho- logic condition; location of pathologic condition; intussusception type; bowel resection requirement; and length of hospital stay.

Data were collected by 2 trained data extractors using a standardized data collection sheet. A sample of 10 records was selected for review by both investigators indepen- dently to assess for agreement. Five key outcome measures (presence of bloody stool, intussusception type, diagnostic test, pathologic condition, and bowel resection requirement) were selected from each of the records for interrater reliability assessment. Reviewers were blinded to each other’s results, and interrater reliability between the reviewers was evaluated using Cohen’s ? statistic.

Study investigators met frequently to discuss data abstraction progress and to resolve any abstraction disputes. To limit data entries to the predefined categories, all data were recorded into a Microsoft Excel 97 database designed with data validation parameters that would not allow extraneous entries (Microsoft Corp, Redmond, Wash).

On the basis of the most common age categories sited in frequently referenced medical texts we reviewed, patients were separated into 3 age groups: traditional pediatric-age, 0 to 6 years old (traditional); nontraditional pediatric-age, older than 6 to 18 years (nontraditional); and adult-age, older than 18 years (adult).

Data analysis

All data elements were entered into a password- protected Microsoft Excel database. Data were analyzed using SPSS for Windows, version 11.0 (Statistical Package for the Social Sciences, Inc, Chicago, Ill). Descriptive statistics was used to describe the demographic character- istics of the study cohort. Nominal variables were described with numbers and percentages, and group comparisons were made using ?2 analysis. Interval level data were described using measures of central tendency. Group comparisons were made using the independent samples t test or analysis of variance, as appropriate for the number of groups. Confidence intervals were calculated by the exact method, and statistical significance was set at an ? of less than .05.

Results

Ninety-five cases of imaging or surgery-proven intussus- ception were diagnosed during the study period: 61 patients in the traditional (64%), 12 in the nontraditional (13%), and 22 in the adult category (23%). The mean age and age range in years for each category were as follows: traditional 1.7 (0.2-5.9), nontraditional 12.6 (8.8-17), and adult 48 (21-91). Interrater agreement between the data abstractors on 5 key data elements represented excellent agreement with ? scores ranging from 0.97 (95% confidence interval, 0.58-1.0) to 1.0 (95% confidence interval, 1.0-1.0).

Overall 19% of patients were seen by a primary care physician for a related abdominal complaint before the diagnosis of intussusception being made: 12 (20%) tradition- al, 3 (25%) nontraditional, and 3 (14%) adult patients. Six percent of these patients were seen multiple times. Similarly, 27% of patients had previously visited an ED for related complaints: 15 (25%) traditional, 5 (41%) nontraditional, and 5 (23%) adult patients. Seven percent of these patients had multiple ED visits relating to the same complaint.

Abdominal pain was the most frequent chief complaint at the time of the final ED presentation: 79% overall, 74% traditional, 92% nontraditional, and 86% adult patients. The average duration of acute pain was 4 hours (range, b1 hour to 4 days), with 30% of patients experiencing similar episodes of pain in the past. An intermittent and cramping pain pattern was described in more than 80% of cases in all groups. Pain severity was categorized as moderate to severe in most patients in all categories: 71% overall, 72% traditional, 83% nontraditional, and 59% adult. Pain location followed no discriminating pattern and was most often described as diffuse.

Vomiting was the second most frequent chief complaint (62% overall, 61% traditional, 91% nontraditional, and 52% adult). Diarrhea occurred in approximately 1 of 4 patients. There was no statistically significant difference among groups for these 2 complaints. The one chief complaint that was significantly different was the presence of bloody stool. This was statistically more common in traditional patients than nontraditional and adult patients: 16% vs 0% vs 4% (P = .016).

Table 3 Associated findings

Traditional Nontraditional Adult All

(n = 61) (n = 12) (n = 22) (n = 95)

Adenitis? 12 (20%) 1 (8%) 1 (5%) 14 (16%)

Meckels 5 (8%) 3 (26%) 0 8 (8%)

diverticulum?

Peutz-Jeghers 1 (2%) 6 (50%) 0 7 (7%) polyp?

Malignancy? 0 1 (8%) 7 (32%) 8 (8%)

Nonspecific 42 (68%) 1 (8%) 3 (13%) 46 (48%)

inflammation?

Other? 1 (2%) a 0 11 (50%) b 12 (13%)

* P b .001, statistically significant difference between the groups.

a Duplication of the distal ileum.

b Prior gastric bypass-associated small bowel intussusception (10), lipoma (1).

Seventy percent of traditional, but only 42% nontradi- tional and 50% adult patients, were admitted with a diagnosis of intussusception (P = .015). Overall, 62% of patients were

Traditional Nontraditional Adult All

(n = 61) (n = 12) (n = 22) (n = 95)

Air contrast 61% 8% 0 40% enema?

Bowel rest? 3% 0 14% 5%

Colonoscopy 0 0 5% 1%

Surgery? 36% 92% 81% 54%

Bowel 23% 75% 43% 34%

resection?

* P b .001, statistically significant difference between the groups.

admitted for symptomatic intussusception. The 2 next most frequent admitting diagnoses were bowel obstruction (7%) and gastroenteritis (6%). Air contrast barium enema (36%) and ultrasonography (33%) were the most common diag- nostic tests in traditional patients, whereas computed tomographic scanning was the test of choice in nontraditional (83%) and adult (68%) patients. Intussusception locations are described in Table 1. Treatments and bowel resection rates are described in Table 2.

Table 2 Treatment type and bowel resection rate

The most frequent associated Pathologic findings for traditional, nontraditional, and adult patients differed signif- icantly and are described in Table 3. Of note, 10 adult patients (11% of the entire cohort) had prior gastric bypass surgery identified as the likely cause of their intussusception. No patient died as a result of their intussusception.

Once adjusted for age, the following indicators proved not to be discriminatory: sex; admission blood pressure, pulse rate, and temperature; pain duration; history of abdominal pain; prior known gastrointestinal condition; known family history of gastrointestinal condition; admission white cell blood count, absolute neutrophil count, hemoglobin level, hematocrit, glucose, blood urea nitrogen, creatinine level, and electrolytes; and length of hospital stay.

Table 1 Location of intussusception

Location Traditional Nontraditional Adult All (n = 61)

93%

Small bowel? 0

Large bowel 7%

(n = 12)

50%

33%

17%

(n = 22) (n = 95)

Ileocolic?

27%

50%

23%

72%

16%

12%

* P b .001, statistically significant difference between the groups.

Discussion

Our case series supports the general teaching that young patients less than 6 years old represent most intussusception cases. However, our findings also suggest that older children and adults are not infrequently susceptible to this same condition. It has previously been reported that intussuscep- tion accounts for 1% to 8% of bowel obstructions in adults [3-6]. Our observations that older children (aged 7-18 years) and adults (N18 years) account for 13% and 23% of symptomatic intussusception, respectively, are, to our knowledge, the highest rates reported to date.

A number of previously published case reports have clearly described the entity of adult intussusception [7-10]. However, 2 reported case series have specifically attempted to more comprehensively define the characteristics of intussusception in the older patient. Eisen and colleagues

[6] performed an 11-year retrospective review of patients 16 years and older who were eventually diagnosed with intussusception at a single institution in New York. They identified 27 patients who demonstrated the following characteristics: median age, 52 years; preoperative diagnosis of intussusception established in 40%; and 22 small bowel and 5 large bowel lesions. Thirty-six percent of the small bowel lesions were malignant (all metastatic) and 80% of large bowel lesions were malignant (all primary adenocarci- nomas). Most patients required bowel resections. They appropriately emphasize the need to specifically identify the primary cause of intussusception in this older patient population given the dire consequences of missing a serious yet possibly curable underlying condition.

Wang and colleagues [5] recently reported a 20-year retrospective review from China. Using the age cutoff of 18, they classified 24 (8.2%) of 292 patients as being adults with acute intussusception. Some of their more pertinent observa- tions include mean age of 49.5 years, bloody stool complaint in 8%, the diagnosis was made preoperatively in 75% of patients, 83% had a defined lesion (55% benign and 45% malignant), and nearly 90% required some type of bowel resection.

The presentation, etiology, and implications of intussus- ception in adults contrast significantly with that of children. Because of its reported rarity, physicians are probably less likely to have a high index of suspicion for acute intussusception in older children and adults presenting with undiagnosed abdominal pain. In addition, adults often lack the classic findings typically seen in children. Rather, their presentation can be vague, intermittent, and chronic, leading to Delays in diagnosis and management with resultant increased morbidity. A significant minority of our patients in all age groups had previously been seen one or more times by either a primary care or emergency medicine provider for similar Abdominal complaints before the diagnosis of acute intussusception being made. Only half of the older patients in our series were admitted with the correct diagnosis of acute intussusception.

Presenting complaints for intussusception in all age groups were predictable as follows: diffuse, intermittent, cramping, and moderate to severe abdominal pain associated with vomiting in most patients. Diarrhea was reported in 1 of

4 patients. The only discriminating complaint between groups was the presence of bloody stool, and this was observed infrequently in only 16% of traditional pediatric-age patients and extremely rarely in the nontraditional pediatric- age and adult-age patients (0% and 4%, respectively). These symptoms and signs are nonspecific and can be caused by a variety of more commonly occurring gastrointestinal condi- tions such as bowel obstruction and gastroenteritis.

Although the etiology of intussusception in children is often either idiopathic or related to self-limited conditions such as mesenteric lymphadenitis, adult intussusception more commonly involves a distinct pathologic Lead point, with up to 50% of previously reported cases attributable to neoplastic disease, either primary or metastatic [5,6,11]. In our series, no traditional pediatric-age patient was discovered to have a gastrointestinal malignancy, whereas 8% of the nontraditional pediatric-age and 32% of the adult-age patients had either a primary or metastatic lesion causing the intussusception. Because of this, adults with intussus- ception usually require surgical reduction or resection and are rarely definitively treated with air contrast barium Enema reduction, which is routinely effective in children. Precisely determining the cause of intussusception is essential in the older patient.

Our finding that 10 (45%) of 22 adult patients had small bowel intussusception associated with a prior gastric bypass procedure reflects recent case reports and case series published in the medical literature [12-20]. This is now an accepted complication of Roux-en-Y Bariatric surgery. Possible causes include disordered intestinal motility, the presence of staple lines and other lead points in the intestinal wall, and the anatomical peculiarities of the enteroenter- ostomy. These are usually small bowel within small bowel intussusceptions, can be either antegrade or retrograde, mimic more common postoperative complications in this patient population, and require a high index of suspicion to make an early diagnosis and avoid serious outcomes.

There are several limitations inherent in our study. As with any retrospective analysis, specific study data elements were not always recorded contemporaneously at the time of actual patient management. In addition, this is a single academic institution experience and might not be transferable to other institutions or geographic locations. Finally, it is always possible that the finding of intussusception in a particular patient was incidental and not related to her complaint of abdominal pain. Factors mitigating this include the substan- tial minority of patients in all groups visiting a primary care or emergency medicine provider one or more times for related complaints before the diagnosis of intussusception being made, the impressive number of patients requiring surgery and bowel resection, and the number of patients with an identifiable likely cause of their intussusception.

Our review would suggest that the incidence of acute intussusception is substantially higher in nontraditional age groups than has been previously reported, with more than 1 in 3 patients in our series lying outside the generally accepted age range. Older patients are less likely to have bloody stool, are less likely to be admitted with a diagnosis of intussusception, are more likely to have associated pathologic conditions of concern, and are more likely to require bowel resection. Gastric bypass surgical procedures appear to be an emerging cause of small bowel intussusception in adults. All adults with intussusception require definitive diagnostic testing to deter- mine the cause, given the concerning list of possibilities.

References

  1. Kleigman RM, Behrman RE, Jenson HB, Stanton BMD, editors. Nelson textbook of pediatrics. 18th ed. Philadelphia (Pa): Saunders; 2007. p. 1569-70.
  2. Marx JA, Hockberger RS, Walls RM, editors. Rosen’s emergency medicine: concepts and clinical practice. 7th ed. Philadelphia (Pa): Mosby; 2010. p. 1185.
  3. Stewardson RH, Bombeck CT, Nyhus LM. Critical operative management of Small bowel obstruction. Ann Surg 1978;187:189-93.
  4. Azar T, Berger DL. Adult intussusception. Ann Sur 1997;226:134-8.
  5. Wang LT, Wu CC, Yu JC, et al. Clinical entity and treatment strategies for adult intussusceptions: 20 years’ experience. Dis Colon Rectum 2007;50:1941-9.
  6. Eisen LK, Cunningham JD, Aufses AH. Intussusception in adults: institutional review. J Am Coll Surg 1999;188:290-5.
  7. Baudendistel TE, Haase AK, Fitzgerald N. Clinical problem-solving. The leading diagnosis–a 23-year-old black woman presented to the

emergency department with diffuse, colicky abdominal pain of 1 hour’s duration. N Engl J Med 2007;357:2389-93.

  1. Yalamarthie S, Smith RC. Adult intussusception: case reports and review of the literature. Postgrad Med J 2005;81:174-7.
  2. Huang BY, Warshauer DM. Adult intussusception: diagnosis and clinical relevance. Radiol Clin N Am 2003;41:1137-51.
  3. Yuzefovich M, Albin JL, Macri CJ. Pediatric emergencies presenting in pregnancy: a report of adult intussusception. The Female Patient 2009;34:41-2.
  4. Nagorney DM, Sarr MG, McIlrath DC. Surgical management of intussusception in the adult. Ann Surg 1981;193:230-6.
  5. Takeuchi K, Tsuzuki Y, Ando T, et al. The diagnosis and treatment of adult intussusception. J Clin Gastroenterol 2003;36:18-21.
  6. Ver Steeg K. Retrograde intussusception following Roux-en-Y gastric bypass. Obes Surg 2006;16:1101-3.
  7. Efthimiou E, Court O, Christou N. Small bowel obstruction due to retrograde intussusception after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2009;19:378-80.
  8. Zainabadi K, Ramanathan R. Intussusception after laparoscopic Roux- en-Y gastric bypass. Obes Surg 2007;17:1619-23.
  9. Coster DD, Sundberg SM, Kermode DS, et al. Small bowel obstruction due to antegrade and retrograde intussusception after gastric bypass: three case reports in two patients, literature review, and recommenda- tions for diagnosis and treatment. Surg Obes Relat Dis 2008;4:69-72.
  10. Edwards MA, Grinbaum R, Ellsmere J, et al. Intussusception after Roux-en-Y gastric bypass for morbid obesity: case report and literature review of rare complication. Surg Obes Relat Dis 2006;2:483-9.
  11. Bocker J, Vasile J, Zager J, et al. Intussusception: an uncommon cause of postoperative small bowel obstruction after gastric bypass. Obes Surg 2004;14:116-9.
  12. Duane TM, Wohlgemuth S, Ruffin K. Intussusception after Roux- en-Y gastric bypass. Am Surg 2000;66:82-4.
  13. Sandrasegaran K, Rajesh A, Lall C, et al. Gastrointestinal complica- tions of bariatric Roux-en-Y gastric bypass surgery. Eur Radiol 2005; 15:254-62.