Secondary Logo

Journal Logo


New Endoscopic and Diagnostic Techniques: Working Group Report of the First World Congress of Pediatric Gastroenterology, Hepatology, and Nutrition: Management of GI Bleeding, Dysplasia Screening, and Endoscopic Training—Issues for the New Millennium

Spolidoro, José Vicente*; Kay, Marsha; Ament, Marvin; Cadranel, Samy; Fujimoto, Takao; Gilger, Mark; Kato, Seiichi; Olives, Jean-Pierre; Goncalves, Manoel Ernesto Pecanha*; Wyllie, Robert

Author Information
Journal of Pediatric Gastroenterology and Nutrition: August 2002 - Volume 35 - Issue - p S196-S204
  • Free

Over the last 10 years, there have been a number of significant advances in the field of pediatric gastrointestinal endoscopy. Pediatric endoscopists have adapted adult endoscopic techniques to the unique spectrum of disease in children and adolescents. Some techniques developed for one indication in adults have been adapted for other uses in children. Three areas that the working group feels will be of increasing importance in the next 10 years include techniques for evaluation and management of GI bleeding, dysplasia and cancer screening in the pediatric patient and acquisition of skills in pediatric GI endoscopy.


I. Summary of the Problem

In the early years of endoscopy, the endoscope was used primarily as a diagnostic tool in patients with GI bleeding of obscure or unknown origin. Before the widespread availability of pediatric endoscopy, antiquated techniques such as nasogastric lavage were used to determine whether a GI bleed was from the upper GI tract and to quantify the severity of the bleed. Endoscopy now plays a role in the evaluation of almost all children with GI bleeding, and bleeding sites from the esophagus to the proximal jejunum and from the distal ileum to the rectum can readily be identified endoscopically in most centers. Some institutions also use push enteroscopy techniques using a pediatric colonoscope in older children and adolescents and an adult gastroscope in younger children to identify and treat lesions of the proximal small intestine.

Therapy of GI bleeding in children was initially limited by endoscope size, and the frequency and natural history of these lesions. Significant bleeding from peptic ulcers was less common in the pediatric age-group because of the lack of co-morbidities such as alcohol consumption, nonsteroidal anti-inflammatory use, and multisystem organ failure. However, nonsteroidals are being increasingly administered to children, and there are an increasing number of children with multisystem organ failure in pediatric intensive care units due to advances in cardiac surgery, cancer treatment, ventilatory support, and transplantation.

II. Major Issues in Need of Investigation or Implementation

Upper GI Bleeding

Treatment of lesions in the upper GI tract can usually be divided into two categories, esophageal and gastroduodenal. Esophageal bleeding in children is usually due to varices, an area in which pediatric endoscopists have a great deal of experience. Most older pediatric endoscopists are comfortable with injection sclerotherapy, a technique that requires knowledge of precise volumes of injection and good endoscope tip control. Over the last several years, endoscopic variceal band ligation has gained increasing popularity in adults. This technique has been shown to be equally efficacious to sclerotherapy, is easier to perform in the setting of acute bleeding, especially for novice endoscopists, and in some series is associated with a lower rate of complications (1,2). Originally ligating devices employed a single-band application device that required reloading with each subsequent band application.

This repetitive esophageal intubation often required or was performed using an overtube. Size constraints of the overtube limited use of this technique in children. Multi-band ligators are now available with four or six bands that can be sequentially deployed, eliminating the need for repetitive esophageal intubation and use of an overtube. This increases the utility of band ligation in older children and adolescents. Some of the bands may be deployed outside of the patient before the procedure, improving the endoscopist's view during the procedure. Current band ligators are still sized for “adult endoscopes” in the 9-mm range and add an additional 2 to 3 mm in outer diameter to the scope. We hope that in the future manufacturers will adapt equipment to allow band ligation using new pediatric video endoscopes with outer diameters in the 6- to 7-mm range to allow more widespread application of this technique in younger children with extrahepatic portal hypertension or biliary atresia and cirrhosis. One disadvantage of esophageal band ligation is that young endoscopists who learn this technique primarily will have less experience with injection as a method of hemostasis in the GI tract.

Injection of sclerosing or hemostatic agents is one of the cornerstones of therapy for children with GI bleeding due to ulcers in the stomach or duodenum (3). This technique, although quite popular with adult endoscopists, has had only limited application in pediatric patients because of a lack of familiarity with the mechanics of the technique. Injection therapy can be performed using almost any size endoscope with a minimum 2-mm working channel, and is very effective for management of bleeding ulcers or preventing rebleeding in the patient with a nonbleeding visible vessel. The limitations with this technique in children and complications reported are primarily attributable to a lack of familiarity with the properties of the different solutions used. Volumes of injection for absolute alcohol-based solutions, for example, are one tenth to one thirtieth those of epinephrine-based solutions, and overinjection can result in bowel necrosis and complications such as gastric perforation. Injection can be performed nontangentially at angles that thermocoagulation would be difficult to perform. Small-diameter metal needles allow for repetitive injection if necessary even in the gastric cardia with the endoscope in the retroflexed position.

Thermocoagulation techniques available for therapy of GI bleeding in children include monopolar and bipolar coagulation, application of the heater probe, laser photocoagulation, and the new argon plasma coagulator. Bipolar coagulators and the heater probe require use of an adult-sized endoscope with a 2.8-mm channel, limiting their application in the upper GI tract in infants and small children. Thermocoagulation using the heater probe, monopolar and bipolar probes rely on the principle of coaptive coagulation, a technique in which the exposed vessel is tamponaded by direct tangential pressure, and then coagulating currents are applied (3). These coagulators are best suited to lesions that can be approached tangentially.

The Argon Plasma Coagulator (A PC) introduced in 1995 uses a noncontact electrocoagulation technique in which high-frequency monopolar alternating current is delivered to tissues through ionized argon gas. The gas is injected through a flexible Teflon fiber. The primary indications currently are for ablation and hemostasis. Lesions amenable to treatment include arteriovenous malformations, gastric antral vascular ectasia, radiation proctitis, polyp bases, and some GI tumors (4). The major advantage of this device over other types of coagulators is the ability to limit the depth of tissue penetration with correct settings and application times, although transmural injury can occur with incorrect use. Laser coagulators and the A PC are designed to be “painted” on lesions, making them suitable for nontangential and noncontact approaches, although the A PC can also be applied tangentially. Although these methods have not found widespread application in pediatric patients, their increasing popularity with adult gastroenterologists and especially the portability of the A PC suggest that these modalities will be increasingly used in children with recurrent bleeding from vascular malformations, and also perhaps in children with polyposis syndromes or radiation injury.

Lower GI Bleeding

Colonoscopy for lower GI bleeding in children is currently used primarily as a diagnostic rather than a therapeutic technique. Colonoscopy can be used to localize bleeding to a site proximal to the ileocecal valve, such as bleeding from a Meckel's diverticulum, although a Meckel's diverticulum is easier to identify using a nuclear scan. Bleeding in the colon is more commonly due to “colitis” than to discrete lesions. However, there are numerous reports of colonic lesions in the pediatric literature that may be amenable to thermocoagulation or injection techniques such as Dieulafoy's lesion of the colon, blue rubber bleb nevus syndrome, colonic varices, and vascular malformations. Injection of saline in the base can be used to raise up a large polyp, making endoscopic removal safer. New detachable polypectomy snares can be applied to a polyp's stalk either before polypectomy or as an adjuvant to therapy for post polypectomy bleeding. Successful band ligation of bleeding lesions in the lower GI tract has also been recently reported. Pediatric endoscopists will need to be familiar with thermocoagulation settings in the lower GI tact, and the appropriate reduction of settings, especially in the right colon.

III. Proposed Plan to Achieve Goals

No large studies evaluate the comparative efficacy of different injection solutions in pediatric patients. It is unlikely that any one pediatric center will accumulate enough pediatric patients with bleeding ulcers to prospectively evaluate any given solution compared with another solution or thermal technique. It may be possible for the pediatric GI community to organize a multicenter study to determine appropriate volumes and solutions for injection in children.

Thermocoagulation techniques in both the upper and lower GI tract in children including watt and joule settings and time of probe or coagulator contact could be evaluated in a similar manner. Although animal studies have been done to determine safe and effective injection volumes and coagulation settings in adults, similar animal studies geared to pediatric patients could give pediatric endoscopists appropriate safety ranges for children. Other hemostatic techniques such as hemostatic clips and injection of fibrin glue will also require evaluation to determine if they are to play a role in pediatric gastrointestinal endoscopy.



The three premalignant gastrointestinal conditions affecting pediatric patients amenable to endoscopic screening are inflammatory bowel disease, Barrett's esophagus (columnar epithelium lined esophagus), and inherited polyposis syndromes including Hereditary Nonpolyposis Colorectal Cancer (HNPCC). Some authors argue that H. pylori infection predisposes to gastric malignancy in adults and untreated celiac disease increases the risk of small bowel lymphoma, but these disorders are not included in this discussion.

Ulcerative Colitis and Crohn Disease

I. Summary of the Problem

Adenocarcinoma of the large bowel associated with ulcerative colitis was first reported by Crohn in 1925. Hospital-based series using actuarial methods of analysis suggested risks as high as 9% to 42% after 25 years of disease. Population-based series suggest the risk of developing adenocarcinoma is 3% to 10% at 25 years in patients with universal colitis (5). The risk of developing cancer is highest in individuals with extensive large bowel disease, and seems to rise after 8 to 10 years of disease. Early reports suggested patients with left-sided colitis had a reduced risk of cancer, but recent reports suggest the incidence of cancer is similar to that in patients with extensive colitis. Isolated ulcerative proctitis does not appear to increase the risk of cancer. Individuals with primary sclerosing cholangitis or cholestatic liver disease and ulcerative colitis may be at increased risk for developing dysplasia compared with patients with ulcerative colitis alone (6). Age of onset is not associated with an increased risk of cancer but younger patients will be at risk for a greater number of years. Folate deficiency may predispose individuals to the risk of development of dysplasia but this remains controversial.

Dysplasia is the neoplastic alteration of the colonic epithelium thought to result from a repetitive cycle of inflammation and regeneration in patients with ulcerative colitis. In 1983 the Inflammatory Bowel Disease Morphology Study Group attempted to standardize the terminology for the assessment of dysplasia (7). Low-grade dysplasia is defined as enlargement of the mucosal crypts, decreased mucin production, nuclear enlargement and hyperchromasia, cellular crowding, and nuclear stratification. High-grade dysplasia has more extreme nuclear changes with pleomorphism and hyperchromasia. In high-grade dysplasia, the nuclear changes are not limited to the basal potion of the crypts but extend toward the surface. Dysplasia may occur on polyps, nodules, plaques, or flat mucosa. Tissue is usually classified as negative, indefinite, or positive for dysplasia. Specimens positive for dysplasia as classified as low or high grade. Differentiation of low-grade dysplasia from regenerative or inflammatory changes may be difficult (7). The latent period from low- to high-grade dysplasia and from high-grade dysplasia to cancer is unknown.

The aim of surveillance is to prevent cancer deaths associated with ulcerative colitis. Analysis of 10 prospective ulcerative colitis dysplasia studies with a total of 1,225 patients showed that when colectomy is performed for high-grade dysplasia, cancer is present in 30% to 40% of the patients (5). In the same series, the probability of finding cancer in colectomy specimens performed with low-grade dysplasia was 8% to 9%. Analysis of four series since 1990 involving 423 patients followed between 1973 and 1990 revealed eight patients with cancer. No cancer deaths occurred during surveillance and 11 patients underwent colectomy for dysplasia. The studies emphasize the importance of initial screening because two patients already had developed carcinoma. Survival rates in patients with colorectal carcinomas detected as part of a surveillance program exceed survival rates in hospitals without surveillance programs and in patients who elect not to participate in screening programs who develop colorectal carcinoma (5).

II. Major Issues in Need of Investigation or Implementation

Ulcerative Colitis.

Current recommendations for screening are that four biopsies should be obtained randomly approximately every 10 cm for a total of 32 or more specimens during colonoscopy. Any suspicious lesions should also undergo biopsy. Screening usually begins at 8 to 10 years after the onset of symptoms, which is often when the pediatric patient is still under the care of their pediatric gastroenterologist, and is repeated every 1 to 3 years. Dysplasia is most often found in the rectum and sigmoid colon. Half the patients with unequivocal low-grade dysplasia will go on to develop high-grade dysplasia or carcinoma within 5 years. The detection of low-grade dysplasia should lead to surgical intervention or more frequent screening (every 3–6 months). An initial finding of low-grade dysplasia with lack of dysplasia identification on subsequent screening procedures may be due to sampling error. Patients should be warned of the inherent risk of cancer and the limitations of screening if conservative monitoring is to be continued. High-grade dysplasia is associated with cancer in 45% of patients and should lead to colectomy. The decision to recommend surgery is tempered by a technical failure rate of ileoanal pouch anastomosis of approximately 5%. In addition, 15% to 50% of the patients will experience at least one episode of pouchitis, and 5% to 10% may suffer from chronic pouchitis. Patients with chronic pouchitis may also be subject to developing dysplasia in the pelvic reservoir, although the frequency and significance of this observation has yet to be established.

Difficulties in identifying individuals at risk has led to a variety of procedures that attempt to augment routine dysplasia screening. The most widely publicized technique has been the determination of aneuploidy by flow cytometry analysis. Aneuploidy is a population of cells with an abnormal amount of DNA. Some studies suggest that aneuploidy identifies a subset of patients who are more likely to develop dysplasia while those without aneuploidy would need less surveillance (8). There remain discrepancies in the literature concerning the usefulness of flow cytometry as an adjunct to routine histologic screening (8). Other promising markers include mucin-associated sialosyl-Tn expression, p53 suppressor gene mutations, and 18q deletion.

Colectomy with ileostomy, ileal rectal anastomosis, or pelvic pouch procedure has been advocated for patients with medically uncontrolled ulcerative colitis or dysplasia. Adenocarcinoma has been reported in the ileostomy site of patients with ulcerative colitis undergoing colectomy, and in patients after ileorectal anastomosis. Dysplasia with neoplastic transformation of the pelvic pouch mucosa has also been reported after ileal pouch anal anastomosis. Patients at highest risk for developing dysplasia after the pelvic pouch procedure are those with a chronic pouchitis and atrophic mucosal changes.

Crohn Disease

An increased incidence of small bowel cancer with a variable risk of large bowel cancer has been observed in population based studies in patient with Crohn disease. Studies from some large referral centers have reported increased risks of large bowel cancer in patients with Crohn disease, whereas other centers found no particular increased occurrence. Patients who may be at increased risk include those with early onset of disease, long-standing anorectal disease, and those with strictures and fistulas (9). Case reports include adolescent age patients but carcinoma is usually seen in older individuals. Surveillance of the large bowel in Crohn disease remains controversial but may be appropriate for younger patients with long-standing colonic disease. Biopsies are performed by using the protocols established for patients with ulcerative colitis. Dysplasia screening protocols have not been established for patient with isolated small bowel disease.

III. Proposed Plan to Achieve Goals

Pediatric gastroenterologists are in the unique position to collaborate with our colleagues in adult gastroenterology and industry to perform long-term outcome studies of endoscopic surveillance starting in adolescent aged patients. These types of studies will help to determine whether there are unique features to pediatric IBD that predispose our patients to malignancy, and whether there are specific dietary or medication interventions that may reduce the risk of developing dysplasia and malignancy. In addition, because pediatric patients are less likely to have other comorbid diseases, it may be easier to assess the utility of various markers and methodologies for detecting dysplasia in our patient age-group.

Familial Adenomatous Polyposis

I. Summary of the Problem

Patients with a family history of familial adenomatous polyposis (FAP), also known as familial polyposis coli (FPC), adenomatous polyposis coli (A PC), or Gardner's syndrome, and patients with HNPCC (formerly known as Lynch syndrome) are at higher risk of developing colorectal and small bowel cancer. Recent developments in genetic testing have enabled earlier detection of affected family members, yet have raised questions as to appropriate methods and timing of screening.

FAP, characterized by an autosomal dominant mode of transmission in two thirds of patients and occurring sporadically in the other one third, can be detected in 80% to 90% of affected individuals who have a relative with a known mutation by genetic testing. The mutation in the tumor suppressor A PC gene is on chromosome 5q21, inactivates the gene and leads to inhibition of apoptosis and proliferation of epithelial cells. Additional tests are available to detect possible carriers in a family with a known proband and include indirect ophthalmoscopy to identify multiple, bilateral, or large areas of congenital hypertrophy of the retinal pigment epithelium (CHRPE). Because of the very high incidence of colon cancer in affected individuals with FAP in early adulthood, endoscopic screening is initiated in late childhood, and carried out through adolescence and young adulthood with the hope of performing colectomy and ileal pouch anal anastomosis in advance of the development of cancer. Current recommendations include annual flexible sigmoidoscopy starting by age 10 to 12 years, annual colonoscopy at age 20 years if attenuated FAP is suspected, prophylactic colectomy usually in the teen years or when polyps are detected at endoscopy, and endoscopic surveillance every 4 to 6 months after ileo-rectal anastomosis and annually after ileoanal anastomosis (10). Family members of an affected individual without a known mutation require regular endoscopic screening. Family members of an affected individual with a known mutation who themselves test negative on genetic testing should still undergo endoscopic screening, although less frequently that individuals with an identified mutation.

II. Major Issues in Need of Investigation or Implementation

It will be the responsibility of the pediatric gastroenterologist to initiate these screening regimens, to perform genetic testing, and to a limited extent with the assistance of genetic counselors to counsel families regarding the results of such testing in their children. The challenge for pediatric gastroenterologists will be to ensure that the correct children are appropriately screened, and that a minimum of children will be unnecessarily screened. It remains to be seen what role nonsteroidal anti-inflammatory drugs such as Sulindac will have in the progression of adenomas in patients with FAP, but as with IBD pediatric gastroenterologists can initiate long-term outcome studies of these types of medications and endoscopic surveillance programs in conjunction with their adult gastroenterology colleagues. Although colectomy will eliminate the risk of colorectal cancer in these patients, pediatric gastroenterologists will also need to counsel and initiate screening protocols in these patients for small bowel cancer (duodenal adenomas, and periampullary tumors), desmoid tumors, pancreatic tumors, hepatoblastoma, thyroid cancers, and other lesions known to be associated with FAP. Duodenal and ampullary cancers are currently the most common cause of cancer deaths in this population, with a risk more than l00-fold that of the general population and occurring in 5% to 10% of patients with FAP (11). Gastric cancer risk is higher in patients with FAP from Japan and Korea but is not increased in patients from Western countries. Current recommendations are to perform upper endoscopy every 6 months to 36 months starting by age 20, but these recommendations will be subject to further review and revision (10).

III. Proposed Plan to Achieve Goals

Multicenter studies will help to determine the appropriate screening intervals and protocols, and collaborative working groups can provide the mechanisms for dissemination of these recommendations. Endoscopic screening protocols and outcome studies can also be developed on a national or international basis for patients affected with other polyposis syndromes with onset in the pediatric age-group with a malignant predisposition, including Peutz-Jeghers's syndrome and Turcot's syndrome.

Hereditary Nonpolyposis Colorectal Cancer

I. Summary of the Problem

Hereditary nonpolyposis colorectal cancer (HNPCC or Lynch syndrome) was originally described in 1913, although further characterization as a family cancer syndrome occurred in the 1960s by Lynch, whose name the syndrome carries. Only recently has the molecular basis of the syndrome been recognized. This autosomal dominant syndrome appears to be due to a mutation in one of several DNA mismatch repair genes, with different mutations identified in different families. This disease is global in distribution, with large series reported from the United States, England, New Zealand, the Netherlands, Italy, Israel, and Finland among other countries (12). HNPCC is thought to account for 1% to 5% of all cases of colorectal cancer (6,000–7,000 cases annually) with a lifetime risk of an affected individual developing colon cancer of 80% (10). Although the average time of diagnosis of cancer in patients with this syndrome is in the mid 40s, cases of cancer have been reported in patients as young as 13 years old. The designation “nonpolyposis colorectal cancer” is a misnomer because cancers arise from adenomatous polyps, which tend to grow and progress rapidly, and frequently occur proximal to the splenic flexure (70% vs. 30% of other colorectal cancers). There is a high rate of synchronous or metachronous colorectal cancers. Patients with this syndrome also have a higher incidence of extracolonic cancers including endometrial cancer, ovarian cancer, gastric and small bowel cancer, and transitional cell cancer of the ureter and renal pelvis. (Lynch II families) There is an excess of poorly differentiated carcinoma, mucinous carcinoma, signet cell carcinoma, and undifferentiated histology in patients with HNPCC (12).

II. Major Issues in Need of Investigation

Currently more than 90% of known mutations in patients with HNPCC are of two genes, hMSH2 (chromosome 2p) and hMLH1 (chromosome 3p). As with FAP, failure to detect a known mutation does not preclude the diagnosis and individuals at risk for HNPCC should undergo appropriate screening (10). Current recommendations for endoscopic screening of family members are colonoscopy every 1 to 3 years beginning at age 20 to 25 or 10 years before the earliest diagnosis of colorectal cancer in the family, total colectomy at the time of diagnosis of cancer, and to consider prophylactic colectomy in known gene carriers. Prophylactic hysterectomy and oophorectomy should be considered in known gene carriers, especially when child bearing is complete or at the time of menopause with careful gynecologic surveillance in the interim.

III. Proposed Plan to Achieve Goals

As with FAP, pediatric gastroenterologists can initiate and evaluate the efficacy of screening protocols for this disorder and counsel families affected with this serious ailment. Multicenter multinational studies would allow for rapid patient accrual, and further understanding of this very prevalent disorder.

Barrett's Esophagus

I. Summary of the Problem

Barrett's esophagus characterized by the presence of specialized columnar epithelium within the tubular esophagus is a preneoplastic condition in children and therefore amenable to endoscopic surveillance. Early studies of the prevalence and outcome of this disorder were hampered by a heterogeneous patient population with the inclusion of patients with several types of abnormal esophageal mucosa not meeting the current pathologic criteria. Only mucosa characterized by goblet cells that stain positive with alcian blue pH 2.5 carries the premalignant risk. A full description of Barrett's is beyond the scope of this working group statement.

II. Major Issues in Need of Investigation or Implementation

Pediatric data suggest an infrequent occurrence of true specialized columnar epithelium, being found in 0.02% to 0.6% of pediatric upper endoscopies from large referral centers (13). Comorbidities in pediatric patients include neurologic impairment, chronic lung disease including cystic fibrosis, esophageal atresia, and malignancies after chemotherapy (13). Adenocarcinoma arising from Barrett's epithelium has been reported in the pediatric age-group. The diagnosis of Barrett's esophagus may require repetitive endoscopic biopsy procedures in the setting of “exudative camouflage” in patients with untreated severe erosive esophagitis. Endoscopic surveillance should be performed every 3 to 4 years in affected patients, starting at age 10 years, if dysplasia is not present. If dysplasia is detected, current protocols should be followed. Any endoscopic protocol requires at least two to four biopsies taken in a stepwise progression at 1- to 2-cm intervals starting 1 cm below the esophagogastric junction and extending 1 cm above the squamocolumnar junction. Multiple biopsies are also required from any abnormal appearing mucosal segment or ulcer. Invasive or intramucosal cancer requires surgical therapy. High-grade dysplasia is frequently associated with carcinoma, and different centers perform either surgery (in younger healthier patients) or more frequent endoscopic surveillance. Indeterminate dysplasia is usually reevaluated after a 4- to 8-week course of aggressive anti-reflux therapy, whereas low-grade dysplasia is usually evaluated by serial surveillance endoscopic biopsies at an initial frequency of every 6 to 12 months. New therapies to treat Barrett's esophagus include use of the argon plasma coagulator discussed earlier, and hopefully prospective studies on the use of this device will be performed in children with this disorder.

III. Proposed Plan to Achieve Goals

As with PAP and HNPCC collaborative studies will provide meaningful long-term outcome data regarding the natural history of this disease and the appropriate endoscopic surveillance intervals.


I. Summary of the Problem

Due to the dramatic growth and developments in the field of pediatric gastroenterology over the past two decades, pediatric endoscopy has evolved from a straightforward diagnostic tool to a field of its own, including diagnostic and therapeutic applications. Training in the field of pediatric endoscopy must keep pace with these developments. It is the responsibility of national and international pediatric gastroenterology organizations to guide its membership in the initial acquisition of skills during fellowship training and the maintenance and enhancement of those skills after completion of training.

Endoscopy in children requires appropriate cognitive and technical skills necessary to diagnose and treat disorders of the gastrointestinal tract, liver, and pancreas. Diagnostic competence is the ability to recognize and understand abnormalities as they are identified. Technical competence is the ability to recognize the need for a diagnostic or therapeutic procedure and the ability to safely and successfully perform the indicated task.

II. Major Issues in Need of Investigation or Implementation

The North American Society for Pediatric Gastroenterology and Nutrition (NASPGN) established guidelines for training in Pediatric Gastroenterology in 1999 (14). The ESPGHAN has also developed guidelines in pediatric endoscopy to fulfill the criteria for recognition by the European Board of Pediatrics, Section of Pediatric Gastroenterology. In these guidelines, two levels of training were recommended. Level 1 trainees are expected to achieve competence in the procedural skills necessary to provide comprehensive care to pediatric patients with gastrointestinal disorders. Level 2 trainees are those who plan to perform specialized endoscopic procedures required for children with complex gastrointestinal problems.

Level 1 (Basic Training)

This level of training assumes competence in General Pediatrics. Training programs should provide a comprehensive and balanced view of the relationship between diagnostic and therapeutic procedures. Trainees should perform endoscopic procedures under the direct observation and guidance of an experienced pediatric gastroenterologist. Trainees should learn:

  1. The ability to recommend endoscopic procedures based on the findings of a personal consultation and recognize indications, contraindications, and diagnostic or therapeutic alternatives for endoscopic procedures
  2. To perform procedures safely, completely and expeditiously
  3. To correctly interpret most endoscopic findings and perform endoscopic intervention as needed
  4. To integrate endoscopic findings into a management plan
  5. To understand risk factors and recognize and manage complications
  6. To recognize personal and procedural limitations and know when to request help
  7. To be familiar with JCAHO or other national organizational guidelines regarding sedation and monitoring for pediatric GI procedures
  8. To identify age and indication appropriate endoscopic equipment
  9. The proper cleaning and maintenance of endoscopic equipment with special attention to infection control (14) measures

Procedures considered “basic” for all trainees include upper endoscopy including foreign body removal and lower endoscopy, including polypectomy (Table 1) (14,15). The number of procedures necessary to achieve competence is extremely variable between individual trainees. The number of needed procedures listed in Table 1 serves only as a guideline for achievement of individual competence. Objective criteria that verify procedural competence do not exist. As such, an experienced training director best determines certification of procedural competence of an individual trainee. It is not possible for all training programs to teach every trainee all endoscopic procedures. However, after completion of their fellowship, novice endoscopists should not perform unsupervised procedures in which they have not been trained. Acquisition of advanced endoscopic skills will require additional training, usually post-fellowship or in a fellowship with special facilities and faculty with expertise in these advanced procedures.

Minimum numbers required to achieve competency in endoscopic procedures in pediatric patients

Level 2 (Advanced Training)

This level of training assumes successful completion of a fellowship in Pediatric Gastroenterology. Trainees may learn diagnostic and therapeutic endoscopic retrograde cholangiopancreatography (ERCP), endoscopic sonography, endoscopic laser therapy, intestinal stent placement, pneumatic dilation of achalasia, endoscopic tumor ablation, coagulation using the APC and endoscopic surgery. This training is usually obtained gradually in collaboration with an experienced endoscopist, often an adult gastroenterologist. Such advanced training necessitates that certain basic requirements are fulfilled. These include:

  1. A program director to assure the following: a structured curriculum that ensures appropriate exposure to children with digestive diseases; competent teachers teaching the technical and cognitive skills; appropriate teaching materials, including books, atlases, videotapes; ongoing review of the trainee's progress to determine competence; ongoing review and update of the training program, including teacher's training materials and facilities.
  2. Appropriate, up-to-date facilities for training in the comprehensive care of children (i.e., inpatient and outpatient facilities, laboratory, pathology and radiology facilities).
  3. Recognition of age- and size-appropriate equipment and training in endoscopic procedures (i.e., recognition of the different needs of an infant vs. an adolescent)
  4. Access to certified specialists in pediatric critical care, surgery, anesthesia, radiology and pathology (14)
  5. The assessment of competence in advanced training requires direct observation of the trainee, ongoing evaluation to correct and improve skills and formal documentation of procedural skills. The number of procedures needed to be considered competent in more advanced endoscopy is detailed in Table 1. These procedure numbers represent only a guideline and the training director must assess each trainee on an individual basis.
Maintenance and Development of New Skills Beyond Fellowship Training

The acquisition of skills in pediatric endoscopy is achieved during fellowship training in Pediatric Gastroenterology. Board certification in Pediatric Gastroenterology verifies the individual's knowledge and skills, although specific evaluation of endoscopic skills is not a separate component of the board certification process. Currently, there is no formal mechanism to ensure the maintenance and development of new skills in pediatric endoscopy.

III. Proposed Plan to Achieve Goals

The First World Congress of Pediatric Gastroenterology, Hepatology and Nutrition should serve as an impetus for the establishment of a program(s) for the maintenance and enhancement of skills in pediatric endoscopy. The working group on new endoscopic and diagnostic techniques proposes that the World Congress and its affiliated Societies develop and support: 1) an annual course(s) on Pediatric Endoscopy; 2) training videos and training centers as part of the annual course in pediatric endoscopy; and 3) Centers of Excellence in Pediatric Endoscopy.

Annual Course(s) on Pediatric Endoscopy

Using the successful “Master Class in Pediatric Endoscopy” by Dr. Michael Thomson as an outstanding example, it is proposed that the World Congress formally develop similar courses, particularly in North America. This course is a 1- to 2-day course that covers all aspects of diagnostic and therapeutic pediatric endoscopy. Other courses could follow the model of the American Society of Gastrointestinal Endoscopy (ASGE) and other organizations that hold small regional “hands-on courses” with a limited number of participants and faculty with expertise in the techniques being evaluated. Courses using animal models have been shown to be particularly effective for initial teaching of therapeutic techniques, and endoscopy simulators will be an important adjuvant to training of both basic and advanced techniques in the next decade.

Training Videos in Pediatric Endoscopy

This simple technology has been successfully developed and employed by the ASGE for many years and offers expert demonstration of endoscopic techniques and image recognition using “video quizzes.” Currently videos are often used in the training of pediatric endoscopists, but are not pediatric specific. Specialized pediatric endoscopy videos could be developed, shown at meetings, placed into digital format and made available on the NASPGN website as “jpeg” programs. Such an arrangement might allow “online” CME in pediatric endoscopy via such web-based programming. These videos could be especially valuable to pediatric endoscopists in solo practice who may not have regular interaction with other pediatric endoscopists and for endoscopists in remote locations worldwide.

Centers of Excellence in Pediatric Endoscopy

It is recognized that there is a limited need for advanced pediatric endoscopy in all medical centers. Only a small number of centers have the appropriate pediatric and adult gastroenterology training programs that are able to offer state-of-the-art endoscopy services for children. It is proposed that several of these centers be designated “Centers of Excellence in Pediatric Endoscopy.” Such designation would include specific funding for the development of the aforementioned proposals. It is hoped that this funding would lead to a national and international cooperative effort to create ongoing “Continuing Medical Education” for pediatric endoscopy.


The working group recognizes that the above discussion and proposals address only a limited number of issues in the field of pediatric gastrointestinal endoscopy. Other areas of potential investigation include optimal methods of sedation, development of new endoscopic instruments and accessories and how they can be used for pediatric patients, new methods of image storage and transmission, the frequency and causes of endoscopic complications, and the comparative yield of endoscopy versus other diagnostic techniques, including “virtual endoscopy.” New techniques such as endoscopic ultrasonography may find increasing application in the pediatric age-group, especially with advances in equipment size. Standard techniques such as percutaneous endoscopic gastrostomy placement may be adapted for neonates and premature infants. It is the hope of the working group that these ideas will stimulate further discussion, research, and collaborative studies in the field of pediatric gastrointestinal endoscopy.


1. Hassall E, Treem WR. To stab or strangle: how best to kill a varix? J Pediatr Gastroenterol Nutr 1995; 20:121–4.
2. Fox VL, Carr-Locke DL, Connors PJ, et al. Endoscopic ligation of esophageal varices in children. J Pediatr Gastroenterol Nutr 1995; 20:202–8.
3. Wyllie R, Kay MH. Therapeutic intervention for non variceal gastrointestinal hemorrhage. J Pediatr Gastroenterol Nutr 1996; 22:123–33.
4. Johanns W, Luis W, Janssen J, et al. Argon plasma coagulation (A PC) in gastroenterology: experimental and clinical experiences. Eur J Gastroenterol Hepatol 1997; 9:581–7.
5. Wyllie R. Dysplasia screening in inflammatory bowel disease. Conference proceedings of the Annual Meeting of the American Society for Gastrointestinal Endoscopy; May 19, 1998, New Orleans, Louisiana.
6. Loftus Jr., EV, Sandborn WJ, Tremaine WJ, et al. Risk ofcolorectal neoplasia in patients with primary sclerosing cholangitis. Gastroenterology 1996; 110:432–40.
7. Riddell RH, Goldman H, Ransohoff DF, et al. Dysplasia in inflammatory bowel disease: standardized classification with provisional clinical applications. Hum Pathol 1983; 14:931–68.
8. Rubin CE, Haggitt RC, Burmer GC, et al. DNA aneuploidy in colonic biopsies predicts future development of dysplasia in ulcerative colitis. Gastroenterology 1992; 103:1611–20.
9. Ekbom A, Helmick C, Zack M, et al. Increased risk of large-bowel cancer in Crohn's disease with colonic involvement. Lancet 1990; 336:357–9.
10. Terdiman JP, Conrad PG, Sleisinger MH. Genetic testing in hereditary colorectal cancer: indications and procedures. Am J Gastroenterol 1999; 94:2344–56.
11. Burke CA, Beck GJ, Church JM, et al. The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic screening program. Gastrointest Endosc 1999; 49:358–64.
12. Lynch HT, Smyrk T. Hereditary nonpolyposis colorectal cancer (Lynch syndrome): an updated review. Cancer 1996; 78:1149–67.
13. Hassall E. Columnar-lined esophagus in children. Gastroenterology Clin North Am 1997; 26:533–48.
14. Rudolph CD, Winter HS, NASPGN Executive Council, et al. NASPGN Guidelines for Training in Pediatric Gastroenterology. J Pediatr Gastroenterol Nutr 1999:29(Suppl 1).
15. Standards of Training Committees, American Society of Gastrointestinal Endoscopy. Principles of training in gastrointestinal endoscopy. Gastrointest Endosc 1999;49:845–53.
© 2002 Lippincott Williams & Wilkins, Inc.