Synopsis  

The pediatric patient is different from the adult one not only in terms of his or her constitutional make-up but also in terms of the pathologies encountered. Although many of the basic principles and techniques are similar, the differences and needs are such that pediatric endoscopy has evolved into a specialized discipline unto itself. This relatively new discipline requires pediatric-specific training, instrumentation, and knowledge in order to accomplish procedures that are safe, comfortable, and ultimately of benefit to the child's well-being.

 Introduction   

Thirteen years after the introduction of the fiberoptic gastroscope by Basil Hirshowitz in 1957 [1] the first reports of flexible gastrointestinal endoscopy in children were published [2–6]. The earliest pediatric endoscopes were either ill-adapted to children or to the gastrointestinal tract itself (e.g. small caliber bronchoscopes). Since that period, an extraordinary evolution has occurred with the advent of progressively thinner and more flexible pediatric-dedicated instruments. Current technology permits safe visualization, tissue sampling, and therapeutic interventions of the upper and lower gastrointestinal tracts in even preterm newborns. It has transformed the practice of pediatric gastroenterology.

The old adage 'children are not small adults' applies especially to the field of pediatric gastroenterology. The pediatric patient must be viewed as an evolving, dynamic patient in terms of anatomy, physiology, and psychological make-up. In addition, pediatric gastroenterologists must be familiar with the multitude of congenital and acquired anomalies not seen in adult practice. This chapter will highlight the differences between the worlds of adult and pediatric endoscopy. It is organized under four major topics:

• the endoscopy facility and personnel;
  
• the pediatric patient and procedural preparation;
  
• endoscopic procedures currently performed in pediatric patients;
  
• selected gastrointestinal pathologies in pediatric patients.
  

 The endoscopy facility and personnel   

Endoscopy facility   

Proper care of pediatric patients requires a child-friendly environment and medical personnel specifically trained in the care of children and adolescents. Pediatric endoscopic procedures, like those performed in adult patients, are routinely performed in a variety of locations. In dedicated pediatric procedural suites and in procedural units shared with adult colleagues, the safe and efficient flow of patients should be assured through the careful organization of the structure and function of these areas.This information, specific to each facility, should be detailed in a policy and procedures manual.

The pediatric endoscopist may also be called to patient areas that are not primarily designed for endoscopy, such as the neonatal or pediatric intensive care units. In these instances, a mobile cart with the necessary equipment, including an endoscope of the appropriate size, a light source and monitor, and a complete panoply of diagnostic and therapeutic instruments together with their generating units (e.g. electrocautery unit), is necessary. The procedure and patient monitoring should be conducted no differently than in a dedicated endoscopy unit.

Equipment   

Endoscopes   

The design of gastrointestinal endoscopes continues to evolve, with video-endoscopes now largely supplanting their older fiberoptic counterparts. Video technology has brought about impressive advances in the field of optics, and innovations such as zoom technology and image processing. The three major manufacturers (Olympus, Fujinon, and Pentax) have pediatric product lines with similar characteristics. Each provides brilliant, high-resolution color views of the gastrointestinal mucosa through a wide angle. The depth of view ranges from 5 to 100 mm, with nearly a 30-fold magnification of the mucosa. Depending on the manufacturer, smaller-sized duodenoscopes, enteroscopes, and variable-stiffness colonoscopes can be found. In addition, adult therapeutic (two-channel) gastroscopes and 'zoom' gastroscopes allowing magnification (up to 150×) of the mucosal image have occasional applications in children.

For most pediatric gastroenterologists, the instrument chosen to perform an endoscopic procedure is based upon personal preference, clinical experience, and level of comfort for both the patient and endoscopist alike. Outer tip diameter and working channel diameter are the two specifications that are of greatest importance in the selection of an endoscope for the smaller child. Outer tip diameters currently range from 5.0 to 12.8 mm in gastroscopes and 11.3 to 13.3 mm in colonoscopes and sigmoidoscopes. Working channel diameters range from 2.0 mm in the slimmest instruments (generally less than 5.9 mm outer diameter) up to 4.2 mm for the largest adult therapeutic endoscopes. Published guidelines for the choice of the appropriate-sized endoscope for a child based on size are few [7,8], perhaps because issues other than size need to be considered. These may include the weight and height of the child, indications for the procedure, and the type of sedation used. For example, a 5 mm gastroscope, passed through a pacifier, has been used in minimally sedated nursing infants by the authors, while we have also used adult-sized endoscopes in ventilated infants (over 6 kg) for therapeutic procedures such as foreign body removal.

Some authors feel that an adult colonoscope may be safely used in patients as small as 15 kg [9]. Because the external diameter of most adult colonoscopes ranges between 13 and 14 mm and they are significantly stiffer than pediatric instruments, they must be used with extreme caution in young patients to minimize the risk of colonic laceration or perforation. Olympus has recently introduced a colonoscope with variable stiffness of the insertion tube, the PCF 160 AI/L. The stiffness of the colonoscope may be adjusted during the procedure by turning a dial on the upper portion of the shaft. This has been shown to decrease looping in the sigmoid and transverse colon upon insertion; however, stiffening of the colonoscope should be performed with caution [10]. Pediatric or adult gastroscopes with outer diameters ranging from 6 to 9 mm may also be substituted for standard colonoscopes for colonoscopy in smaller patients.

Endoscopy instruments   

Adult endoscopic instruments are appropriate for children if an adult-sized endoscope is being used. Many of these instruments have been redesigned to pass through the smaller diameter (2.0 mm) working channels of the infant-sized endoscopes. These 5F-sized instruments include: biopsy forceps, grasping forceps (e.g. rat-toothed, 'penny pinchers', alligator, etc.), polypectomy snares, injection catheters, baskets, cautery probes (for argon plasma coagulation), and dilation guidewires. Currently lacking are band ligation devices, electrocautery probes, balloon dilators, and foreign body hoods.

Ancillary equipment   

In addition to the necessary light sources, monitors, mobile carts, and monitoring equipment, many endoscopy units have access to software allowing image processing and electronic reporting of the endoscopic procedure. Photo-documentation of the procedure can be helpful from both a clinical and liability standpoint, and whenever possible should be appended to the endoscopic record. Software allowing reporting of an electronic endoscopic record is now available from both commercial and non-commercial sources. In particular, PEDS-CORI (http://www.peds-cori.org) is a non-commercial venture that supplies an electronic pediatric endoscopy-report software program in exchange for collecting outcomes data on endoscopic procedures in children. This banking of information is allowing important outcome studies to be performed. Regardless of the source, these programs should allow accurate and complete recording of the procedure as well as generate data for billing and quality improvement purposes. It is imperative that they be adapted to the pediatric practice.

Personnel   

The endoscopist   

'Endoscopy in children requires appropriate cognitive and technical skills necessary to diagnose and treat disorders of the gastrointestinal tract, liver and pancreas'[11]. Professional societies in both Europe and North America have set forth processes that will allow trainees in pediatric gastroenterology to acquire these skills [11,12]. These processes assume competence in general pediatrics and completion of a fellowship in which two levels of endoscopic competency can be achieved: basic and advanced. Minimum numbers of supervised procedures have been described and procedural proficiency tools are currently under development to help the trainee achieve competency. Guidelines for the maintenance of acquired skills and development of new skills beyond training are also being discussed.

Pediatric practice guidelines have also been issued for adult endoscopists [13]. Adult endoscopists may be called upon by pediatric primary care providers to render basic services when a pediatric gastroenterologist is not available, or to provide advanced services not routinely performed by a pediatric gastroenterologist (e.g. ERCP). Although the indications for endoscopy in children may be similar to those for adults, a team approach is recommended whenever possible to avoid potentially unnecessary or even deleterious procedures.

Any gastroenterologist performing pediatric procedures should have thorough knowledge of the pediatric airway and be certified in PALS (Pediatric Advanced Life Support, offered by the American Heart Association/American Academy of Pediatrics) unless assisted by a qualified anesthesiologist.

Nursing and ancillary personnel   

Nurses and endoscopy technicians play an integral role in pediatric endoscopy. They must be specifically trained and comfortable in the field of endoscopy and in meeting the needs of children. In addition, they must be adept at communicating with families and patients, with specific attention to the developmental and cognitive levels of the latter. Safe and efficient endoscopy, when utilizing conscious sedation, is best performed with the help of two assistants. One assistant monitors the patient during the procedure while the other assists the endoscopist with tissue sampling and therapeutic interventions. The assistant assigned to monitoring the patient continually assesses the patient's vital signs and comfort level, can physically and emotionally provide support for the patient, monitors patient positioning, and can give the endoscopist periodic updates on the patient's degree of abdominal distension. When general anesthesia is employed, only one endoscopy assistant is usually necessary to assist with biopsies and therapeutic interventions.

Most institutions require that nursing personnel involved in sedated pediatric procedures maintain skills and certification in PALS (see above). These highly specialized procedural nurses now have their own professional organizations through which more information can be obtained (http://www.naspgn.org/sub/apgnn.asp and http://www.sgna.org).

 The pediatric patient and procedural preparation   

Patient preparation   

Preparation of the pediatric patient for an endoscopic procedure requires two equally important components: psychological preparation and medical preparation.

Psychological preparation   

It has been well demonstrated that appropriate preparation prior to an endoscopic procedure will reduce patient distress and anxiety, thereby reducing overall use of sedative medications and enhancing patient safety [14,15]. Information regarding the need for the procedure, how and where it will be performed, and the process of recovery is given to both the patient and parents. Information given should be honest and accurate, and that communicated to the child should be done in a sensitive manner appropriate to his or her developmental level. Ideally, this is done by the endoscopist well in advance of the procedure and away from the procedural unit.

Often, innovative techniques are used to help prepare the child. For example, placement of an IV catheter for sedation is especially anxiogenic for children [14]. It may be reduced by using a preprocedural nursing unit where the catheter is placed by skilled pediatric phlebotomists after application of a topical anesthetic such as EMLA™. Furthermore, a 'cooling off' period after the IV has been placed has been found to be helpful. Additional techniques used to reduce preprocedural anxiety include the use of illustrative pictures and dolls, preparatory videos, diagrams, visits to the endoscopy suite, and distribution of written materials. Explanatory pamphlets can be downloaded from http://www.NASPGHAN.org and many pediatric units have produced their own personalized guides. These approaches require additional time, effort, and personnel but are ultimately found to be helpful.

Medical preparation   

This begins with a careful history, physical examination, and assessment of patient ASA classification [16] (Fig. 1). These will determine the type of procedure needed, its appropriateness, procedure location (e.g. operating or endoscopy suite), personnel and equipment support needed, and level of sedation required. In addition, laboratory assays, imaging tests, and assessment of parental or patient expectations may be helpful in making these decisions.

Recommendations for fasting   

The need for preprocedural fasting continues to be an evolving issue. Traditionally, and based upon little data, children undergoing a sedated endoscopic procedure have been required to fast for a minimum of 8 h. This practice is uncomfortable for infants, young children, and parents alike. It can also lead to dehydration in very young infants and hypoglycemic events in children with underlying metabolic disorders. A body of evidence now supports more lenient fasting guidelines [17–19] and is supported by the American Society of Anesthesiologists [20] (Fig. 2). These guidelines should be followed for any child who is to receive sedation and/or undergo an esophagogastroduodenoscopy (EGD).

Bowel preparation   

Bowel preparation for lower intestinal endoscopy has been well standardized for adults [21]. It is usually based upon a large volume lavage with oral polyethylene glycol (PEG) in an electrolyte-balanced solution. Pediatric gastroenterologists have long recognized that these protocols fail miserably in most children due to the poor taste and large volumes of solution required. Some children may be hospitalized and a nasogastric tube placed for instillation, but this adds an extra degree of cost and discomfort that is most often unacceptable. Consequently, most institutions use a variety of agents combined into 'home-grown recipes' (Fig 3). Few of these have been studied and compared. Consensus on the best colonic lavage preparations in pediatric patients has yet to be reached [22–25].

Antibiotic prophylaxis   

The American Heart Association and the American Society for Gastrointestinal Endoscopy have published guidelines for antibiotic prophylaxis prior to endoscopic procedures [26]. The overall risk of bacterial endocarditis as a direct result of an endoscopic procedure is small. Transient bacteremia may occur during or immediately following endoscopy; however, the organisms typically identified are unlikely to cause endocarditis. The rate of bacteremia associated with routine upper endoscopy, colonoscopy, or sigmoidoscopy is between 2 and 5%, and does not increase with mucosal biopsy or polypectomy. Thus, endocarditis prophylaxis is not routinely recommended for 'negligible risk' patients undergoing these endoscopic procedures with or without mucosal biopsy and/or polypectomy. In addition, antibiotics are not recommended for immunocompromised patients or those with pacemakers or prosthetic joints. Other endoscopic procedures such as varix sclerosis, stricture dilation, and ERCP can be associated with bacteremia in up to 45% [27]. High-risk patients include those with prosthetic heart valves, a prior history of endocarditis, a surgically constructed or congenital systemic–pulmonary shunt, or a synthetic vascular graft that is less than 1 year old. These patients should receive antibiotic prophylaxis prior to all endoscopic procedures associated with increased risk of bacteremia, as mentioned. In these patients, however, there are not sufficient data to make recommendations regarding antibiotic prophylaxis prior to routine endoscopy with mucosal biopsy or polypectomy, and the decision is left with the patient's endoscopist and cardiologist, to decide on a case-by-case basis.

Recommended antibiotics and dosages for bacterial endocarditis prophylaxis are as follows: ampicillin 50 mg/kg IM/IV (not to exceed 2.0 g) and gentamicin 1.5 mg/kg IV/IM (not to exceed 120 mg), given 30 min prior to the procedure; 6 h later, ampicillin 25 mg/kg IM/IV or amoxicillin 25 mg/kg po is given. If the patient is penicillin-allergic, vancomycin is administered in place of the ampicillin and is used in combination with the aforementioned dose of gentamicin. Vancomycin 20 mg/kg IV (not to exceed 1 g) is administered over 1–2 h and completed within 30 min of starting the endoscopic procedure [26].

It should also be noted that all children undergoing percutaneous endoscopic gastrostomy tube placement should be given prophylactic antibiotics to decrease the risk of soft tissue infection from skin pathogens.

Informed consent   

As with other invasive procedures, it is mandatory to obtain informed consent prior to endoscopic procedures in pediatric patients according to institutional guidelines. If the patient is under the age of 18 years, the custodial parent or legal guardian must render the informed consent. Although they are unable to provide informed consent, the procedure should also be explained to the patient in an age-appropriate manner.

 Endoscopic procedures currently performed in pediatric patients   

Indications and limitations   

Indications for gastrointestinal endoscopy and colonoscopy in pediatric patients are summarized in Figs 4 and 5. They are adapted from a 1996 medical position statement issued by NASPGHAN (North American Society for Pediatric Gastroenterology, Hepatology and Nutrition) [28]. The primary indications for upper endoscopy in pediatric patients include evaluation of gastrointestinal bleeding, recurrent epigastric pain, nausea or vomiting, feeding or growth abnormalities, and caustic or foreign body ingestion. Additional indications are listed in Fig. 4.

The primary indications for colonoscopy in pediatric patients include gastrointestinal blood loss, unexplained chronic diarrhea, abdominal pain with systemic symptoms suspicious for inflammatory bowel disease (IBD), and dysplasia screening in patients with IBD or polyposis syndromes. Additional indications are listed in Fig. 5.

In contrast to adult patients, colonoscopy is typically not indicated in the pediatric patient with constipation or chronic abdominal pain without other clinical 'red flags' in the history, examination, or laboratory findings, or a 'change in bowel habits', as the incidence of colorectal cancer is exceedingly low in children without underlying predisposing conditions. In addition, a follow-up colonoscopy is not indicated for an asymptomatic pediatric patient found to have fewer than five juvenile polyps.

Absolute and relative contraindications for both upper and lower endoscopies are listed in Fig. 6 and Fig. 7, respectively. It should be noted that children with connective tissue disorders such as Ehler–Danlos syndrome are at increased risk of perforation [29]. Endoscopy should always be avoided when similar results can be obtained in a less invasive manner. Although not contraindicated, endoscopy is generally not helpful in patients with hypertrophic pyloric stenosis or uncomplicated gastroesophageal reflux disease, or for pre-emptive control of varices prior to the first episode of bleeding. Similarly, colonoscopy is not indicated in the setting of acute self-limited diarrhea and may only serve to confuse the clinical picture.

The main limitation in pediatric endoscopy continues to be size, both as it relates to the patient and to instrumentation. As manufacturers continue to develop smaller and smaller endoscopes and their associated instruments, neonatal endoscopic pathology is being described with increasing frequency [30–33]. Gastroscopes with an outer diameter of 5.0 mm have been used for diagnostic endoscopy in children weighing 900 g [32], and for percutaneous endoscopic gastrostomy (PEG) placement in children approaching 2100 g (personal experience, JSE).

Patient sedation   

In pediatric patients, the need for an endoscopic procedure is often synonymous with the need for sedation. The patient-specific goals of sedation include anxiolysis, analgesia, and amnesia for the procedure. Endoscopist-specific goals for sedation include patient safety, adequate endoscopic examination and therapeutic intervention, equipment safety, time efficiency, and patient cost [34]. As the types and complexity of pediatric endoscopic procedures have increased, the type of sedation (general anesthesia, deep, moderate, and mild sedation) has become a more complex issue for the pediatric gastroenterologist [35].

The best method for achieving procedural sedation remains the subject of ongoing debates and is under constant evolution. This is perhaps due to the variety of sedation modalities available, and the complexity and number of issues that need to be taken into consideration when choosing the most appropriate sedation protocol (Fig. 8).

Occasionally, no sedation is administered to pediatric patients undergoing endoscopic procedures. Sedation has traditionally been withheld in very young infants undergoing upper endoscopy, perhaps due to our poor knowledge base of the pharmacokinetics of sedative agents in these patients. This position is becoming increasingly hard to defend [36,37]. Sedation practices in the very young are changing, but require further study [38].

Recently, unsedated upper endoscopies were studied in 21 motivated and consenting children over 8 years of age with good outcomes in terms of tolerance and safety [39]. These patients, however, constituted a very small proportion of the total patients undergoing EGD in that institution. Most pediatric gastroenterologists recognize other instances when sedation may not be required (e.g. small infants or mature children undergoing flexible sigmoidoscopy).

The use of sedation and analgesia, however, remains the norm for most pediatric endoscopic procedures. The choice of an appropriate sedation protocol is complex and takes into account the patient's age, maturity level, disease state, the type and expected duration of the procedure, the risks and benefits of each sedation protocol, and the wishes of the patient and patient's family. Other important issues that enter the decision-making process are outlined in Fig. 8. Consequently, pediatric endoscopists must remain knowledgeable and flexible in their abilities to design different sedation protocols that will meet the many needs of their patients.

A variety of agents can be used by physicians trained in their administration. These physicians should be competent in dose titration to achieve the expected level of sedation, the monitoring of pediatric patients, and the ability to recognize and treat complications as they arise. The choice of appropriate agents should take into account their ability to provide sedation, analgesia, and amnesia. Detailed descriptions of these agents and guidelines for their use go beyond the scope of this chapter and have been well described elsewhere [16,35].

Most endoscopists will use the combination of a narcotic (e.g. fentanyl or meperidine) and a benzodiazepine (e.g. midazolam) to achieve a state of moderate sedation. Recently 'non-traditional' agents for pediatric endoscopy have been described, including propofol [40,41], ketamine [42,43], and sevoflurane [44]. Although effective, they often achieve deep sedative states or can be responsible for specific airway issues requiring monitored anesthesia care for safe usage.

Endoscopic technique   

The overall technique of pediatric endoscopy is similar to that used in adults, but significant differences exist and must be kept in mind. The following is not intended to be an in-depth description of pediatric endoscopic technique, as excellent descriptions can be found elsewhere [9,45]. Rather, it will highlight the differences between pediatric and adult practice. The reader is reminded that any written description of endoscopic technique can only complement and should not replace a hands-on, well-mentored training program.

Esophagogastroduodenoscopy   

Esophagogastroduodenoscopy (EGD) is the most common endoscopic procedure performed by pediatric gastroenterologists. In experienced hands, pediatric upper endoscopy is a safe and minimally invasive procedure. As in adults, EGD in children is often an outpatient procedure and typically takes 15–20 min to complete, depending on procedural findings. The advent of ultra-thin endoscopes, such as the Olympus N30 endoscope with an external diameter of 5.3 mm, allows examination of premature infants as small as 1.5 kg [46].

Preparation for upper endoscopy may differ somewhat in infants. Bite blocks are usually not required in edentulous infants. Nursing infants may use the endoscope itself as a pacifier or the instrument may be passed through a formula bottle nipple in which the tip has been cut away. Also, it has been our practice to avoid the use of topical anesthetic sprays in the pharynx of young infants. We have found its application distressing and usually not worth the small benefit.

Intubation of the esophagus under direct visualization is preferred over blind intubation. It is more comfortable in children receiving moderate levels of sedation but importantly, allows examination of the hypopharyngeal structures to look for signs of gastroesophageal reflux disease, foreign bodies, and laryngeal malformations.

Applying a short puff of air to the hypopharynx may facilitate intubation. This will often produce a cough followed by a swallow which opens the esophageal introitus. The esophagus, from the cricopharyngeus to the thoracic inlet, is narrow and difficult to examine, especially in infants. It is the most frequent site of foreign body entrapment. The distance in centimeters from the gum line to the gastroesophageal junction should be noted in addition to the location of the gastric Z-line in relation to the gastroesophageal junction.

After aspiration of gastric secretions and insufflation until there is near complete effacement of the gastric rugae, the endoscope is advanced through the body of the stomach and into the antrum, carefully observing the mucosa. It should be noted that the junction between the body and antrum, forming the incisura, often makes a more acute angulation than in adults and requires filling the greater curvature of the stomach with the endoscope to allow passage. This places some degree of pressure on the greater curvature that the patient may experience as discomfort.

Retroflexion views of the cardioesophageal junction are a necessary component of the procedure but often difficult in small children, especially when a larger endoscope is used. This maneuver, which can also be performed after examination of the duodenum, requires withdrawal of the tip to the incisura. The up-dial is then turned maximally, often combined with either action of the right/left dial and/or torquing of the instrument. The tip is then withdrawn proximally with frequent twisting of the instrument to provide 360° views of the cardioesophageal junction.

Passage through the pylorus should not be forceful as it will cause the tip of the endoscope to abruptly abut against the distal wall of the duodenal bulb, placing the pediatric patient at increased risk of duodenal hematoma or perforation. This is due to the relatively short pediatric bulb and release of tension on the endoscope that was caused by loading of the greater curvature. After examination of the bulb, the endoscope is passed into the second portion of the duodenum using the same maneuvers as for an adult. Greater care should be taken, however, due to the exiguity of this area. The distal duodenum can then be examined using a straightening maneuver (i.e. a short withdrawal of the endoscope resulting in a paradoxical advancement of the tip).

Biopsies are usually taken upon withdrawal of the instrument. The quality of tissue for histologic evaluation is determined by the size of the biopsy specimen and the choice of location for collection [47,48].

In general the largest possible biopsy forceps for the working channel is chosen. The choice of location for biopsy collection is more problematic. Biopsies should be taken in areas that are visually abnormal or in areas where a suspected diagnosis is most likely to be found (e.g. proximal esophagus in suspected eosinophilic esophagitis).

Even when the procedure is visually normal, it is still advisable to obtain biopsies to confirm normality at a histologic level. In these instances it has been recommended that samples be obtained from the distal duodenum, antrum, or prepyloric area, the body of the stomach along the greater curvature, and in the distal esophagus 2–5 cm above the Z-line [47].

Pediatric gastroenterologists tend to be generally more liberal with mucosal biopsies due to poor correlation between endoscopic and histological abnormalities [49,50]. For example, it is routine to biopsy the distal duodenum in pediatric patients to rule out celiac disease, as creases or notches in the valvulae conniventes are oftentimes not apparent until adolescence.

Upon withdrawal from the stomach, remaining gastric air should be aspirated to facilitate breathing and prevent gaseous discomfort. At the conclusion of the procedure, the abdomen should be manually examined for excessive distension, crepitus, or tenderness.

Colonoscopy   

Understanding the anatomy of the colon and its potential variations is essential for repeated performance of safe and efficient colonoscopic examinations. The degree of difficulty encountered during the examination is in large part determined during fetal development. In 15% of individuals, the ascending and descending colon do not become fixed retroperitoneally in the paravertebral gutters. Instead, they remain free on mesocolons identical to those that are normal for the transverse and sigmoid colon [51]. This results in a potentially mobile colon that may move unpredictably within the abdomen during the procedure. Such anatomy may promote loop formation.

The thickness of the colonic wall and the length of the colon are also key considerations during colonoscopy. The colonic wall is extremely thin, varying in thickness from only 1.7 to 2.2 mm. At birth, the colonic length measures approximately 50 cm, reaching a typical adult length of 90–120 cm in late adolescence.

During colonoscopy to the cecum, this translates into an instrument insertion length of 40–50 cm in young children and 60–80 cm in adolescents and adults once all loops have been successfully removed. Careful monitoring of insertion depth throughout the procedure is even more important in pediatric gastroenterology, especially since the introduction of longer instruments. An insertion length that is significantly beyond the expected colonic length typically signals the presence of a large loop, which may increase the risk of colonic perforation.

Prior to starting the procedure, the height of the examination table should be positioned at the waist level of the endoscopist to optimize control of the colonoscope. In addition, some pediatric patients may require the ambient room temperature to be increased to maintain the patient's core body temperature. This is especially true of young anesthetized infants undergoing a prolonged colonoscopy.

Patient positioning often begins in the left lateral decubitus position, unless general anesthesia is used, in which case a supine, frog-legged position may facilitate the anesthesiologist's care. The procedure should always begin with a careful perianal and digital rectal exam. The presence of perianal disease such as skin tags or fissures, in the absence of constipation, should be considered a sign of inflammatory bowel disease until proven otherwise.

The digital rectal exam, in addition to providing valuable clinical information (e.g. the presence of a stricture, polyp, etc.) will also give an indication as to the adequacy of sedation and bowel preparation. A brief examination of the patient's abdomen prior to the beginning of the procedure is also essential to assess for baseline abdominal distension and tone.

The colonoscope should be advanced through the entire colon, keeping the following general recommendations in mind [45,52]:

• The colonic lumen should be visualized at all times, recognizing that this may not be always possible, at least for short times and distances (e.g. passage of the rectosigmoid junction or splenic flexure).
  
• Minimizing air insufflation will facilitate passage while lessening loop formation.
  
• Loop formation should be avoided whenever possible; loops should be reduced as soon as they are formed.
  
• Frequent short advancements and withdrawals of the colonoscope will allow the colon to 'accordion' onto the instrument, again facilitating passage and the avoidance of loop formation.
  

Correct colonoscopic technique can often be judged by a child who remains comfortable throughout the procedure, a cecum that is reached in most children and adolescents according to the distances described above, and lack of mucosal trauma seen upon withdrawal.

Examination of the rectal mucosa in infants and small children may reveal many small submucosal nodules. They tend to efface with insufflation but may not disappear completely. These nodules represent lymphonodular hyperplasia and are not pathologic (see below).

Endoscopic landmarks, such as the blue silhouettes of the liver and spleen or the triangular contour of the transverse colon, may not always be apparent in very young children. Radiologic confirmation of position, however, is rarely indicated. Verifying successful cecal intubation can be achieved by identifying appropriate structures (e.g. the appendiceal orifice, the ileocecal valve, or the characteristic triradiate haustral folds of the cecal bas-fond), visualization of the instrument's light over the right lower abdominal quadrant, and observing a sharp cecal indentation upon palpation with the index finger in this same area [9].

Intubation of the ileocecal valve and examination of the terminal ileum are important maneuvers. It is especially useful in the evaluation of IBD and unexplained hematochezia. Intubation of the ileocecal valve is perhaps one of the most difficult colonoscopic skills to acquire but is always facilitated by straightening of the instrument. Both indirect [45] and direct visualization [9] intubation techniques have been described. To the adult endoscopist, the terminal ileum of the child often appears abnormal, carpeted by a multitude of submucosal nodules (Fig. 9). This again is a normal finding represented by an exuberance of Peyer's patches. This lymphonodular hyperplasia tends to decrease with age, but may be present even in older adolescents. Occasionally, it may mimic Crohn's disease on small bowel radiologic images. Endoscopically with Crohn's disease, overlying exudates and/or ulceration will also be seen. (Fig. 10)

Distension of the colonic lumen with air allows examination of the entire mucosal surface upon withdrawal of the instrument. It is also upon withdrawal that tissue sampling should be obtained following the same principles as described above [47,48]. Even when the mucosa is visually normal it is still advisable to collect biopsies. It is not unusual, even in experienced hands, to have discrepancies between the visual and histologic findings. Random biopsies are usually taken from the cecum, the transverse, descending, and sigmoid colons, and the rectum, and should be placed in separate containers. Retroflexion views of the distal rectum should be performed when possible prior to final withdrawal of the instrument.

Sigmoidoscopy   

The techniques used for successful sigmoidoscopy are the same as for colonoscopy, recognizing that only the distal large intestine requires examination. Consequently, the required degree of bowel preparation and sedation may be less. In nursing infants, for example, clear liquids for 6–8 h and a formula bottle during the procedure may be all that is needed. If there is a probability that the sigmoidoscopy findings are likely to result in a second procedure (e.g. finding a juvenile polyp in a toddler with hematochezia) then a full colonoscopy should be performed first. This will avoid a second bowel preparation and sedation.

Therapeutic endoscopy   

Therapeutic endoscopy has become a required component of the pediatric gastroenterologist's skills. It has the ability to supplant traditional surgical techniques requiring more invasive and time-needy approaches. Almost all therapeutic techniques, with the exception of percutaneous endoscopic gastrostomy (PEG) placement [53], originated from the adult experience and have been adapted to children. Consequently, general techniques, indications, and contraindications for pediatric therapeutic endoscopy are the same as for adults with, however, several limitations. These limitations include the size of the child, but especially the availability of the specialized instruments that fit through the smaller working channels of the pediatric endoscopes (see above). A list of therapeutic modalities that have been adapted to children are given in Figs 4 and 5.

Other endoscopic modalities   

A number of additional specialized endoscopic modalities need to be presented. Although they are indicated less frequently than the basic procedures described above or fit into the realm of emerging technologies, they nonetheless have become part of the pediatric gastroenterologist's armamentarium. Because the numbers of these procedures are small, the issue of who is best able to perform them continues to be discussed. It is clear that several pediatric referral centres see enough volume to ensure ongoing competency, but in most institutions the procedures require the assistance of trained adult colleagues.

Small bowel enteroscopy   

There is an increasing need and interest in being able to examine the entire small bowel mucosa. This is especially valuable in cases of obscure intestinal bleeding, Crohn's disease, and polyposis syndromes. Three modalities have been devised to examine endoscopically the small bowel: sonde enteroscopy, push enteroscopy, and intraoperative assisted push enteroscopy. Sonde enteroscopy, a passive, per nasal technique, is being largely abandoned for newer, quicker techniques with limited therapeutic capabilities. Push enteroscopy involves peroral passage of a long flexible endoscope beyond the ligament of Treitz. In experienced hands 120–180 cm of small bowel can be examined [9,54]. Pediatric colonoscopes or specialized small bowel enteroscopes may be used. Limitations include increased procedural duration and discomfort, often requiring the use of general anesthetics, and the angulation of the duodenum. This acute angulation dissipates the propelling forces put on the endoscope, resulting in a large and uncomfortable gastric loop [45]. This can be overcome by use of a straightening overtube or newer, variable-stiffness enteroscopes. In addition to visual examination of the mucosa, biopsies may be taken. Therapeutic options are otherwise limited. Intraoperative assisted enteroscopy involves both a surgeon and an endoscopist. It allows for broader therapeutic options, including multiple polypectomies in children with familial polyposis syndromes and ablation of multiple intestinal vascular ectasias. In this procedure the peritoneal cavity is approached either laparoscopically or through an open incisional technique. The small bowel is intubated perorally using an enteroscope, and passage to the terminal ileum is facilitated by the surgeon who 'accordions' the small bowel onto the endoscope as it is advanced.

The procedure is further facilitated by placing a non-crushing clamp at the ileocecal valve and using minimal air insufflation. The mucosa is examined upon withdrawal of the instrument. Air insufflation can be used at this point, while always keeping the intestinal clamp 20–40 cm distal to the tip of the endoscope. It is at this time also that biopsies may be taken, lesions ablated, polyps snared and removed, or the mucosa tattooed for later localization at surgical enterotomy.

Wireless capsule endoscopy   

Since its first description in 2000 [55] this has rapidly become a valuable means to diagnose small bowel mucosal disease. This technology is based upon complementary metal oxide image sensors, application-specific integrated circuit devices, and white light-emitting diode illumination. Its use is now well described in adults [56] and its value to pediatrics is beginning to be studied [57,58]. The video-endoscopic capsule is of a size (11 × 26 mm) such that it may be swallowed by cooperative children.

Although lower-size limits in children have not been published, we have had children weighing 25 kg swallow the capsule without difficulty. In small children, it has been suggested that the capsule may be front-loaded, using a standard retrieval basket, onto the end of an endoscope and deposited into the duodenum using conventional endoscopic technique. The capsule is propelled through the small bowel by peristalsis. The short focal length lens (1 mm) obviates the need for air insufflation and provides excellent images. Video images are radio-transmitted continuously over a 6 h period to a recording device. The stored images are then downloaded to a desktop computer for analysis.

This procedure is indicated for the study of obscure gastrointestinal bleeding, Crohn's disease, and polyposis syndromes, with other indications to follow (e.g. surveillance of intestinal rejection or graft-versus-host-disease after transplantation).

It is contraindicated in patients with known or suspected obstruction or strictures. Its limitations include an inability to obtain tissue or fluid samples, a lack of therapeutic capabilities, and an inability to 'steer' the capsule. Additionally, it requires approximately 1.5–2 h of physician time to adequately review and analyze the stored data.

Endoscopic ultrasonography   

Although well implemented in adult practice, perhaps due to the greater needs for cancer diagnosis, endoscopic ultrasound has yet to be routinely employed in the study of pediatric gastrointestinal disease. Published studies are few [59–61], but demonstrate its usefulness in the study of digestive tumors, angiomatosis, and biliary, pancreatic, and proctologic disease.

Endoscopic retrograde cholangiopancreatography (ERCP)   

ERCP continues to be a highly specialized procedure for the evaluation and treatment of both adult and pediatric pancreatico-biliary disease. This procedure is discussed in another chapter.

 Selected gastrointestinal pathologies in pediatric patients   

Eosinophilic esophagitis   

Eosinophilic esophagitis (EE) is becoming increasingly recognized in pediatric patients. Reflux esophagitis and EE may have similar endoscopic appearances such as circumferential rings and vertical grooves. The presence of white specks adherent to the esophageal mucosa has recently been found to be highly specific for EE [62] (Fig. 11).

The specks have been found to microscopically contain eosinophils. It has been recommended that, in order to distinguish EE from reflux esophagitis, two biopsies should be obtained from the distal esophagus approximately 3 cm proximal to the squamocolumnar junction, and an additional two biopsies from the mid-esophagus.

The recommendation is based on the premise that reflux esophagitis is worse in the distal esophagus, while EE is a more diffuse process [63]. The distinction is critical clinically, as many patients with EE benefit from food allergy testing with subsequent elimination diets and topical corticosteroid therapy, as opposed to acid suppressants therapy alone.

Food allergic enteropathy and colitis   

Gastrointestinal food allergy has gained recognition as a significant pediatric problem over the past several decades. Prospective studies looking at cow's milk protein allergy have reported a prevalence of 2.0% in cow's milk formula-fed infants while breastfed infants demonstrated an incidence of only 0.5% [64].

This is the most frequent form of food allergy during infancy. Typical symptoms include emesis, diarrhea, hematochezia, failure to thrive, malabsorption, and anaemia. Symptom onset may be at birth or delayed for months. There is a cross reactivity with soy protein in 40–60% of infants, thus necessitating use of a cow's milk protein hydrolysate formula or synthetic amino acid-based formula as therapy. The diagnosis is primarily based on clinical history.

Flexible sigmoidoscopy shows mucosal erythema, nodularity, and aphthous ulcers of the rectosigmoid region (Fig. 12). Typical biopsies show eosinophilic infiltrates. Patients typically respond well to dietary elimination programs.

Foreign body ingestion   

This is a common problem faced by pediatric gastroenterologists. Over 80% of all ingestions occur in pediatric patients, with the majority being under the age of 3 years [65]. Symptoms associated with ingestion vary depending upon location of the foreign body and the age of the patient. They include dysphagia, odynophagia, chest pain, choking, drooling, feeding difficulty, respiratory difficulty, abdominal pain, and hematochezia. It is not uncommon that a foreign body is found incidentally on X-rays taken for other reasons such as cough or wheezing.

As many as 90% of ingested foreign bodies in pediatric patients are radiopaque; therefore, an X-ray is recommended in all suspected cases [68]. Of the foreign bodies that are brought to medical attention, up to 90% pass spontaneously, 10–20% require endoscopic removal, and less than 1% require surgical intervention [67].

Coins are by far the most common foreign body ingested by pediatric patients, with pennies ranking number one in the United States [66]. The most problematic coins are those trapped in the esophagus. It is important to note that coins in the esophagus will assume an en face appearance on the AP view while the edge of the coin is seen on a lateral film (Fig. 13).

If the opposite configuration is present, the coin is most likely in the trachea. Patients with a coin in the esophagus who are unable to swallow their secretions or those who are in respiratory distress require emergent endoscopic removal.

Coins caught in the distal esophagus have a high likelihood of spontaneous passage. In these instances a trial of glucagon or conservative management allowing the child to drink, with a repeat X-ray 12–24 h later, may be tried.

Coins trapped at the level of the thoracic inlet will not in most instances pass spontaneously and endoscopic removal should be scheduled within 24 h to minimize the risk of such complications as stricture, tracheal compression, pseudodiverticulosis, tracheo-/bronchoesophageal fistula, aortoesophageal fistula, or esophageal perforation. The endoscopic appearance of a coin in the esophagus is shown in Fig. 14.

We do not routinely recommend removal of coins distal to the gastroesophageal junction unless the patient is symptomatic or the coin has been retained for more than 4–6 weeks, in which case it is unlikely to pass spontaneously.

Although unlikely in adults, battery ingestions are common in pediatric patients. Button batteries are more commonly ingested than their cylindrical counterparts. Significant morbidity can occur if the battery lodges in the esophagus, making this clinical scenario an endoscopic emergency. Reported complications include delayed stricture development, esophageal perforation, fistula development, and even death [69].

Esophageal perforation may occur if the battery remains in the esophagus for as little as 6 h. Thus, rapid and precise distinction between button battery and coin ingestions must be made on X-ray. Button batteries have a 'double-density' appearance, as there is a step-off between the anode and cathode (Fig. 15). If the patient is asymptomatic and the battery is located in the stomach, observation is recommended.

If the battery remains in the stomach for over 48 h or if the patient becomes symptomatic at anytime, we recommend endoscopic retrieval. Such retrieval may be achieved with use of a basket, polypectomy snare, or forceps (tripod or pentapod) device. Cathartics are sometimes beneficial in hastening intestinal transit of the ingested battery. Meticulous screening of the patient's stool or repeat imaging is necessary to confirm passage.

Meat impactions and ingestions of sharp objects are treated the same in pediatric patients as they are in adults. A technique recently developed for management of objects not amenable to removal with standard forceps and baskets (Fig. 16) uses a string or suture placed endoscopically [70] (Fig. 17).

This technique has worked well in the removal of hair barrettes and washers, and should work well for other objects with a hole, such as a toothbrush if the bristles are not embedded in mucosa.

Helicobacter pylori gastritis   

H. pylori infection is the most common cause of gastritis in pediatric patients, and recently appears to be associated with most primary ulcers in children over the age of 10 years [71,72]. Ulcers in children under the age of 10 years tend to be associated with stress events or medications such as corticosteroids or non-steroidal anti-inflammatory medications. Infection during childhood with this organism has also been identified as a risk factor for mucosa-associated lymphoid tissue (MALT) lymphoma and gastric adenocarcinoma in adults. Transmission of the organism is by the fecal–oral and possibly oral–oral routes. Horizontal transmission within families has also been reported [73,74].

The prevalence of H. pylori appears to increase with age. Identified risk factors for infection include poor sanitation, crowding, lower socioeconomic status, poor nutrition, and a family history of peptic ulcer disease [75].

Presenting symptoms of H. pylori infection in pediatric patients includes upper abdominal pain and hematemesis. The upper endoscopy with biopsy remains the preferred diagnostic test for H. pylori in children. The sensitivity of breath urease testing in this population ranges from 75% to 100% in some reports, but is much less sensitive and specific in children under the age of 2 years [76,77].

The endoscopic appearance of the gastric and duodenal mucosa often correlates poorly with the presence or absence of histologic abnormalities in children. We therefore recommend that biopsies be obtained from the gastric antrum, gastric body, and duodenum in pediatric patients undergoing upper endoscopy.

Recently, diffuse antral nodularity has been found to be highly specific for H. pylori gastritis [78]. The characteristic histologic appearance shows chronic gastritis with the presence of mucosal lymphoid aggregates. H. pylori organisms are typically identified on the luminal surface or adherent to the apical surface of mucus cells in modified Giemsa stains, well-prepared hematoxylin-eosin stains, cresyl violet stains, or silver stains.

Triple therapy regimens, similar to those used in adults but modified according to patient weight, have been shown to eradicate the organisms in 70–95% of cases and have documented similar resolution of histologic and clinical abnormalities [79].

There remains, however, significant controversy surrounding the diagnosis and treatment of H. pylori in pediatric patients with non-ulcer dyspepsia. Repeat endoscopy to confirm eradication is generally not indicated unless the patient continues to be symptomatic and indirect diagnostic testing (e.g. stool antigen assay or C₁₃ urea breath test) is negative.

Polyps in the pediatric patient   

Polypectomy is the most common therapeutic application in both adult and pediatric patients undergoing colonoscopy. In contrast to adults, however, the vast majority of polyps in pediatric patients are juvenile polyps with negligible malignant potential unless they are part of a polyposis syndrome. Juvenile polyps account for more than 90% of polyps in children. They occur in approximately 1% of preschool and school age children [80]. This contrasts to adults where the incidence of adenomas is significantly greater. Juvenile polyps may be solitary or multiple, and typically present with painless rectal bleeding, recurrent abdominal pain, or intussusception. Because up to 60% of juvenile polyps are located proximal to the sigmoid colon, pediatric patients undergo a complete colonoscopy rather than flexible sigmoidoscopy when a polyp is clinically suspected [81]. Juvenile polyps may be sessile, but are often pedunculated and can measure up to several centimeters in length and diameter (Fig. 18).

Although the risk of developing malignancy in a solitary juvenile polyp is very small, discovery warrants removal. A single colonoscopy with polypectomy is considered adequate treatment if the lesion is solitary and there is no family history of juvenile polyposis syndrome (Fig. 19).

If new symptoms arise, however, the patient needs reinvestigation. When indicated by the family history or when multiple polyps have been identified, updated routine screening guidelines must be followed.

Adenomas, too, may be found in pediatric patients, but are fortunately rare. They are most often encountered in families with a history of familial adenomatous polyposis (FAP) or hereditary non-polyposis colorectal cancer syndrome (HNPCC). The presence of a colonic adenoma in a patient under the age of 30 years should raise the suspicion for an inherited polyposis syndrome.

Appropriate genetic testing and counseling should be performed in such patients. An APC gene mutation may be detected in 60–80% of index cases.

For at-risk relatives with a negative gene test, FAP is excluded and the individual is considered to have an average population risk for the development of adenomas and colorectal cancer. A positive test in at-risk relatives confirms the diagnosis of FAP and patients should then undergo endoscopic assessment [82].

Annual flexible sigmoidoscopy is recommended from the age of 10–14 years until adenomas are found. Sigmoidoscopy is felt to be adequate because there is early rectal involvement in almost all patients. A complete colonoscopy is recommended by age 16 years to determine polyp load and location as well as the degree of dysplasia [83]. This information is used in combination with the psychosocial and educational needs of the patient to help make an informed decision regarding the timing and type of surgery.

In families in which the genotype is not known, protocols vary. We recommend annual sigmoidoscopy starting at age 10–14 years until rectal adenomas are identified, followed by routine colonoscopic surveillance as above. Some centres recommend that if no polyps are identified by age 20 years, the patient should then start undergoing colonoscopy with dye spray every 5 years [83].

Currently, the most common cause of cancer deaths in patients with FAP is duodenal and ampullary malignancies. Upper endoscopic surveillance of the stomach, duodenum, and periampullary region with a side-viewing scope is not routinely recommended for pediatric patients and should begin after age 20 years. If a pediatric patient with FAP develops unexplained upper abdominal pain, however, earlier investigation by an endoscopist trained in the use of side-viewing scopes is warranted [11].

Lymphonodular hyperplasia   

Lymphonodular hyperplasia is a relatively common lymphoproliferative condition found in pediatric patients at the time of colonoscopy. It is typically most pronounced in the terminal ileum of pediatric patients, but is also commonly seen in the rectosigmoid or scattered throughout the colon. It does not appear to be related to any specific disease process. The submucosal lymphoid follicles typically range in size from 2 to 4 mm.

The condition may be responsible for intermittent hematochezia if located in the rectum or sigmoid. These nodules, either dense or sparsely distributed, usually tend towards effacement with air insufflation of the bowel. Gradual resolution with age is the rule. Stool softeners may help to decrease the amount of blood seen in the stool.

 Outstanding issues and future directions   

Pediatric gastrointestinal endoscopy continues to evolve favorably. Issues approached in this chapter requiring further investigation include:

• improved protocols for pediatric procedural preparation and sedation;
  
• evidence-based protocols for large bowel preparation in children;
  
• expanding the panoply of pediatric-dedicated endoscopy equipment;
  
• ongoing standardization of protocols for training and maintenance of proficiency to avoid the lack of a directed approach to the application of emerging technologies in pediatric gastrointestinal disease.
  

Future directions and initiatives should strive to create unified research and educational agendas that will address these issues and others as they arise. They can be established through the existing infrastructure of professional societies and institutions dedicated to these interests.

 References  

1 Hirshowitz, BI, Peters, CW & Curtiss, LE. Preliminary reports on a long fiberscope for examination of the stomach and duodenum. University Mich Med Bull 1957; 23: 178–80.

2 Kawai, K, Murakami, K & Misak, F. Endoscopical observations on gastric ulcers in teenagers. Endoscopy 1970; 2: 206–8.

3 Ottenjann, R. Gastroscopic extraction of a foreign body. Endoscopy 1970; 3: 193–4.

4 Freeman, NV. Clinical evaluation of the fiberoptic bronchoscope for pediatric endoscopy. J Ped Surg 1973; 8: 213–20.

5 Gleason, PD, Tedesco, FJ & Keating, WA. Fiberoptic gastrointestinal endoscopy in infants and children. J Pediatr 1974; 85: 810–13. PubMed CrossRef

6 Cremer, M, Peeters, JP & Emonts, J et al. Fiberendoscopy of the gastrointestinal tract in children: experience with newly designed fiberscopes. Endoscopy 1974; 6: 186–9.

7 Fox, VL. Pediatric endoscopy. Gastrointest Endosc Clin N Am 2000; 10: 175–94. PubMed

8 Seidman, EG. Role of endoscopy in inflammatory bowel disease. Gastrointest Endosc Clin N Am 2001; 11: 641–57. PubMed

9 Thomson, M. Colonoscopy and enteroscopy. Gastrointest Endosc Clin N Am 2001; 11: 603–39. PubMed

10 Brooker, J, Saunders, B & Shah, S et al. A new variable stiffness colonoscope makes colonoscopy easier: a randomized controlled trial. Gut 2000; 46: 801. PubMed

11 Spolidoro, JV, Kay, M & Ament, M et al. 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. J Pediatr Gastroenterol Nutr 2002; 35: S196–S204. PubMed CrossRef

12 Hassal, E. NASPGN Position Paper: requirements for training to ensure competence of endoscopists performing invasive procedures in children. J Pediatr Gastroenterol Nutr 1997; 24: 345–7. PubMed CrossRef

13 Eisen, GM, Chutkan, R & Goldstein, JL et al. Modification in endoscopic practice for pediatric practice. Gastrointest Endosc 2000; 52: 838–42. PubMed

14 Lewis Claar, R, Walker, LS & Barnard, JA. Children's knowledge, anticipatory anxiety, procedural distress, and recall of esophagogastroduodenoscopy. J Pediatr Gastroenterol Nutr 2002; 34: 68–72. PubMed CrossRef

15 Mahajan, L, Wyllie, R & Steffen, R et al. The effects of a psychological preparation program on anxiety in children and adolescents undergoing gastrointestinal endoscopy. J Pediatr Gastroenterol Nutr 1998; 27 (2): 161–5. PubMed CrossRef

16 Vasundhara, T, Peters, J & Gilger, M. Sedation for pediatric endoscopic procedures. J Pediatr Gastroenterol Nutr 2000; 30: 477–85. PubMed CrossRef

17 Ingebo, KR, Rayhorn, NJ & Hecht, RM et al. Sedation in children: adequacy of two-hour fasting. J Pediatr 1997; 131: 155–8. PubMed CrossRef

18 Schreiner, MS, Triebwasser, A & Keon, TP. Ingestion of liquids compared with preoperative fasting in pediatric outpatients. Anesthesiology 1990; 72: 593–7. PubMed

19 Gleghorn, EE. Preoperative fasting: you don't have to be cruel to be kind. J Pediatr 1997; 131: 12–13. PubMed

20 American Society of Anesthesiologists task force on sedation and analgesia by non-anesthesiologists. Practice guidelines for sedation and analgesia by non-anesthesiologists. Anesthesiology 2002; 96: 1004–17. PubMed CrossRef

21 Faigel, DO, Eisen, GM & Baron, TH et al. Guidelines: preparation of patients for GI endoscopy. Gastrointest Endosc 2003; 57: 446–50. PubMed CrossRef

22 Gremse, DA, Sacks, AI & Raines, S. Comparison of oral sodium phosphate to polyethylene glycol-based solution for bowel preparation for colonoscopy in children. J Pediatr Gastroenterol Nutr 1996; 23: 586–90. PubMed CrossRef

23 Abubakar, K, Goggin, N & Gormally, S et al. Preparing the bowel for colonoscopy. Arch Dis Child 1995; 73: 459–61. PubMed

24 Dashan, A, Lin, C-H & Peters, J et al. A randomized, prospective, study to evaluate the efficacy and acceptance of three bowel preparations for colonoscopy in children. Am J Gastroenterol 1999; 94: 3497–501. PubMed CrossRef

25 da Silva, MM, Briars, GL & Patrick, MK et al. Colonoscopy preparation in children: safety, efficacy and tolerance of high- versus low-volume cleansing methods. J Pediatr Gastroenterol Nutr 1997; 24: 33–7. PubMed CrossRef

26 Dajuni, AS, Taubert, KA & Wilson, W et al. Prevention of bacterial condocarditis: recommendations by the American Heart Association. JAMA 1997; 277: 1794–801. PubMed CrossRef

27 Neu, HC & Fleischer, D. Controversies, dilemmas, and dialogues: recommendations for antibiotic prophylaxis before endoscopy. Am J Gastroenterol 1989; 84: 1488. PubMed

28 Squires, RH & Colletti, RB. Indications for pediatric gastrointestinal endoscopy: a medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr 1996; 23: 107–10. PubMed CrossRef

29 Stillman, AE, Painter, R & Hollister, DW. Ehler–Danlos syndrome type IV: diagnosis and therapy of associated bowel perforation. Am J Gastroenterol 1991; 86: 360–2. PubMed

30 Lazzaroni, M, Petrillo, M & Tomaghi, R et al. Upper GI bleeding in healthy full-term infants: a case-control study. Am J Gastroenterol 2002; 97: 89–94. PubMed CrossRef

31 Ojala, R, Ruuska, T & Karikoski, R et al. Gastroesophageal endoscopic findings and gastrointestinal symptoms in preterm neonates with and without perinatal indomethacin exposure. J Pediatr Gastroenterol Nutr 2001; 32: 182–8. PubMed CrossRef

32 Ruuska, T, Fell, JM & Bisset, WM et al. Neonatal and infantile upper gastrointestinal endoscopy using a new small diameter fibreoptic gastroscope. J Pediatr Gastroenterol Nutr 1996; 23: 604–8. PubMed CrossRef

33 deBoissieu, D, Dupont, C & Barbet, JP. Distinct features of upper gastrointestinal endoscopy in the newborn. J Pediatr Gastroenterol Nutr 1994; 18: 334–8. PubMed

34 Nowicki, MJ & Vaughn, CA. Sedation and anesthesia in children for endoscopy. Techniques in Gastrointestinal Endosc 4 (4), 2002: 225–30.

35 Kaplan, RF & Yang, CI. Sedation and analgesia in pediatric patients for procedures outside the operating room. Anesthiol Clin N Am 2002; 20: 181–94. CrossRef

36 Walco, GA, Cassidy, RC & Schecter, NL. Pain, hurt and harm: the ethics of pain control in infants and children. N Engl J Med 1994; 331: 541–4. PubMed CrossRef

37 Deboer, SL & Peterson, LV. Sedation for nonemergent neonatal intubation. Neonatal Netw 2001; 20: 19–23. PubMed

38 Ng, E, Taddio, A & Ohlsson, A. Intravenous midazolam infusion for sedation of infants in the neonatal intensive care unit. Cochrane Database Syst Rev 2003: (1): CD002052.

39 Bishop, PR, Nowicki, MJ & May, WL et al. Unsedated upper endoscopy in children. Gastrointest Endosc 2002; 55: 624–30. PubMed

40 Elitsur, Y. Propofol sedation for endoscopic procedures in children. Endoscopy 2000; 32: 788–91. PubMed CrossRef

41 Kaddu, R, Bhattacharya, D & Metriyakool, K et al. Propofol compared with general anesthesia for pediatric GI endoscopy: is propofol better? Gastrointest Endosc 2002; 55: 27–32. PubMed CrossRef

42 Law, AK, Ng, DK & Chan, KK. Use of intra muscular ketamine for endoscopy sedation in children. Pediatr Int 2003; 45: 180–5. PubMed CrossRef

43 Hostetler, MA & Barnard, JA. Removal of foreign bodies in the pediatric ED: is ketamine an option. Am J Emerg Med 2002; 20: 96–8. PubMed CrossRef

44 Montes, RG. Deep sedation with inhaled sevoflurane for pediatric outpatient gastrointestinal endoscopy. J Pediatr Gastroenterol Nutr 2000; 31: 41–6. PubMed CrossRef

45 Wyllie and Kay chapter on technique. ?.

46 Fox, VL & Walker, WA et al. (2000) Upper gastrointestinal endoscopy in pediatric gastrointestinal disease. Pathophysiology/Diagnosis/Management (3/e) pp 1401–14 BC Decker Inc. Hamilton Ontario.

47 Gillett, P & Hassall, E. Pediatric gastrointestinal mucosal biopsy. Gastrointest Endosc Clin N Am 2000; 10: 669–711. PubMed

48 Gillett, P. Pediatric gastrointestinal mucosal biopsy: special considerations in children. Gastrointest Endosc Clin N Am 2000; 10: 669–712. PubMed

49 Boyle, JT. Gastroesophageal reflux in the pediatric patient. Gastrointest Endosc Clin N Am 1989; 18: 317–37.

50 Dahms, BB & Rothstein, FC. Mucosal biopsy of the esophagus in children. Perspect Pediatr Pathol 1987: 97–123.

51 Saunders, BP, Phillips, RKS & Williams, CB. Intraoperative measurement of colonic anatomy and attachment with relevance to colonoscopy. Br J Surg 1995; 82: 1491–3. PubMed

52 Cotton, PB & Williams, CB. (1980) Practical Gastrointestinal Endoscopy. pp 99–141. Oxford, Blackwell.

53 Gauderer, MWL, Ponsky, JL & Izant, RJ. Gastrostomy without laparatomy: a percutaneous endoscopic technique. J Ped Surg 1980; 15: 872–5.

54 Perez-Cuadrado, E, Macenlle, R & Iglesias, J et al. Usefulness of oral push video enteroscopy in Crohn's disease. Endoscopy 1997; 29: 745–7. PubMed

55 Iddan, G, Meron, G, Glukhovsky, A & Swain, P. Wireless capsule endoscopy. Nature 2000; 405: 417. PubMed CrossRef

56 Ginsberg, GG, Barkun, AN & Bosco, JJ et al. Wireless capsule endoscopy: August 2002. Gastrointest Endosc 2002; 56: 621–4. PubMed CrossRef

57 Seidman, EG. Wireless capsule video-endoscopy: an odyssey beyond the end of the scope. J Pediatr Gastroenter Nutr 2002; 34: 333–4. CrossRef

58 Mallet, E, Cron, J & Stoller, J. Wireless capsule video-endoscopy: preliminary results in children. Arch Pediatr 2003; 10: 244–5. PubMed CrossRef

59 Kato, S, Fujita, N & Shibuya, H et al. Endoscopic ultrasonography in a child with chronic pancreatitis. Acta Paediatr Jpn 1993; 35: 151–3. PubMed

60 Roseau, G, Palazzo, L & Dumontier, I et al. Endoscopic ultrasonography in the evaluation of pediatric digestive diseases: preliminary results. Endoscopy 1998; 30: 477–81. PubMed

61 Nadler, EP, Novikov, A & Landzberg, BR et al. The use of endoscopic ultrasound in the diagnosis of solid pseudopapillary tumors of the pancreas in children. J Pediatr Surg 2002; 37: 1370–3. PubMed CrossRef

62 Lim, JR, Gupta, SK & Fitzgerald, JF et al. White specks in esophageal mucosa (WSEM): a true endoscopic manifestation of severe eosinophilic esophagitis (EE) in children? J Pediatr Gastroenterol Nutr 2001; 33: 411 (Abstract 164).

63 Liacouras, CA & Wenner, WJ. The quantity of esophageal eosinophils, not the location, is diagnostic of eosinophilic esophagitis in children. Gastrointest Endosc 2000: 51: AB133 (Abstract 3709).

64 Host, A & Halken, S. A prospective study of cow milk allergy in Danish infants during first three years of life. Allergy 1990; 45: 587–96. PubMed

65 Webb, WA. Management of foreign bodies of the upper gastrointestinal tract. Gastroenterology 1988; 94: 204–16. PubMed

66 Jefferson, S. A thought for your pennies. JAMA 1999; 28: 122. CrossRef

67 Arana, A, Hauser, B & Hachimi-Idrissi, S et al. Management of ingested foreign bodies in childhood and review of the literature. Eur J Pediatr 2001; 160: 468–72. PubMed CrossRef

68 Macpherson, RI, Hill, JG & Othersen, HB et al. Esophageal foreign bodies in children: diagnosis, treatment and complications. Am J Roentgenol 1996; 166: 919–24.

69 Litovitz, T & Schmitz, BF. Ingestion of cylindrical and button batteries: an analysis of 2,382 cases. Pediatrics 1992; 89: 747–57. PubMed

70 Kay, M & Wyllie, R. Techniques of foreign body removal in infants and children. Techniques in Gastrointest Endosc 2002: 4 (4): 188–95.

71 Drumm, B, Sherman, P & Cutz, E et al. Association of Campylobacter pylori on the gastric mucosa with antral gastritis in children. N Engl J Med 1987; 316: 1557–61. PubMed

72 Gormally, SM, Kierce, BM & Daly, LE et al. Gastric metaplasia and duodenal ulcer disease in children infected by Helicobacter pylori. Gut 1996; 38: 513–17. PubMed

73 Elitsur, Y, Adkins, L & Saeed, D et al. Helicobacter pylori antibody profile in household members of children with H. pylori infection. J Clin Gastroenterol 1999; 29: 178–82. PubMed CrossRef

74 Mitchell, JD, Mitchell, HM & Tobias, V. Acute Helicobacter pylori infection in an infant associated with gastric ulceration and serologic evidence of intra-familial transmission. Am J Gastroenterol 1992; 87: 382–6. PubMed

75 Hassall, E & Dimmick, JE. Unique features of Helicobacter pylori disease in children. Dig Dis Sci 1991; 36: 417–23. PubMed CrossRef

76 Rowland, M et al. Carbon 13-labeled urea breath test for the diagnosis of Helicobacter pylori infection in children. J Pediatr 1997; 131: 815–20. PubMed CrossRef

77 Madani, S, Rabah, R & Tolia, V. Diagnosis of Helicobacter pylori infection from antral biopsies in the pediatric patient: is urease test that reliable? Dig Dis Sci 2000; 45: 1233–7. PubMed CrossRef

78 Elitsur, Y, Raghuverra, A & Sadat, T et al. Is gastric nodularity a sign for gastric inflammation associated with Helicobacter pylori infection in children? J Clin Gastroenterol 2000; 30: 286–8. PubMed CrossRef

79 Dohil, R, Israel, DM & Hassall, E. Effective 2 week therapy for Helicobacter pylori disease in children. Am J Gastroenterol 1998; 92: 244–7S.

80 Session, R et al. Carcinoma of the colon in the first two decades of life. Ann Surg 1965; 162: 279. PubMed

81 Mestre, JR. The changing pattern of juvenile polyps. Am J Gastroenterol 1996; 81: 312–14.

82 Peterson, GM, Francomano, C & Kinzler, K et al. Presymptomatic direct detection of adenomatous polyposis coli (APC) gene mutations in familial adenomatous polyposis. Hum Genet 1993; 91: 307–11. PubMed

83 Hyer, W. Polyposis syndromes: pediatric implications. Gastrointest Endosc Clin N Am 2001; 11 (4): 659–82. PubMed

Copyright © Blackwell Publishing, 2003