Do Children With Gastroesophageal Reflux Become Adults With Gastroesophageal Reflux? What Is the Role of Acid Suppression in Children?

We covered a number of questions in our committee: Do children with gastroesophageal reflux (GER) become adults with GER? The answer is not known (Table 1).TABLE 1: Do children with GER become adults with GER?Why should we study the natural history of GERD? Adults with reflux often give a history of having had problems that suggest reflux in childhood. These observational data suggest that GER in adults may be a disorder that begins in childhood. Is there a genetic predisposition for GER? Gastroesophageal reflux disorder (GERD) seems to occur more frequently in some families. Genetic studies are just beginning to be done. Could the natural history of reflux be modified by intervention? Data in adults suggest that GER does affect quality of life. So if natural history could be altered in childhood, morbidity in adults might decrease. The development of Barrett's esophagus and the transformation into esophageal adenocarcinoma might be prevented. Could the natural history of extraesophageal manifestations of reflux be altered? If reflux were a confounding variable with asthma or chronic hoarseness, effective therapy might alter the clinical outcome. What is the best way to evaluate GER? Expertise is needed in framing reliable questions for a cohort-type study. One approach to answering these questions might be to “tag on” to a current cohort study, such as the Framingham Offspring Study, the National Longitudinal Survey of Youth, or the Asthma Cohort Study in Tucson, Arizona. These are up-and-running cohort studies that might yield retrospective data, because they have been asking about specific symptoms relevant to GERD. Alternatively, the studies could be modified to add these studies by into their regular reviews. Another approach would be to start a new cohort study, but this strategy requires significant expertise and expense. QUESTIONS AND ANSWERS Q: Should a reflux registry be established, or can we tag on to an existing cohort study? A: There may be some opportunities with existing cohorts. The one that comes to mind for pediatrics would be the asthma work that is being done. We could look at the subset of asthma sufferers who have GER. A: There are problems with many aspects of this issue. Unlike with cystic fibrosis or inflammatory bowel disease, we do not have a standardized diagnostic definition of GERD. Until there is a precise definition, it will be challenging to design multicenter cohort studies. A: Connecting with an existing database that already collects information seems feasible. One of the advantages of being part of the Framingham Offspring Cohort, for example, is that they have done extensive data collection already on food frequency, alcohol intake, and the like. Many factors that have been associated with GERD are already part of their database. Q: Do you have any sense of the severity of GER in adults? Is this a mild illness? A: We surveyed a thousand patients in a managed care organization in the Boston area. These individuals have been receiving chronic proton pump inhibitor (PPI) or histamine-2 (H2) blocker therapy for more than 3 months, usually more than 2 years continuously. We were surprised at the severity of the symptom burden. We found that 30% to 40% of patients who are on high-dose PPIs have differences in their Medical Outcomes Study Short-Form 12 Health Survey (SF-12) quality-of-life indicators. This is an impressive number. People are missing work because of GERD, and they are being treated. The large symptom burden does provide the kind of outcomes that would be possible through a survey-based tool. A: The population-based survey published by Frank et al. (Dig Dis Sci 2000;45:809) looked at quality of life as well. Both patients with GERD and those with “GERD-plus,” or dysmotility-type symptoms in addition to typical GERD symptoms, had a quality of life that was inferior in comparison to control subjects. Clearly there is a high prevalence of impaired quality of life. A: Three different studies have now looked at patients drawn from the general population. These are not health-care-seeking individuals, but people who are taking over-the-counter remedies. In one case, people were identified at the gate of a factory and asked a simple question, “Do you take antacids?” The studies showed that a substantial portion of these individuals have erosive disease. In those who do not have erosive disease, ambulatory pH monitoring showed a significant proportion with nonerosive GERD. So, in the adult, the end point of heartburn is an excellent surrogate for real disease. A: We strongly feel that a focus group be developed to formulate the question before getting involved with established cohort groups. It is important to form a focus group that formulates questions that validate the issues. A: The Pediatric CORI Project and the Adult CORI Project could prospectively include information about reflux on a long-term basis. The CORI Project was started in 1997 by David Lieberman at University of Oregon in the Veterans Administration. CORI, or Clinical Outcomes Research Initiative, is essentially an endoscopic repository, which has now logged 250,000 procedures over the past 3 years. The PEDS-CORI Project, which is funded through the Children's Digestive Health and Nutrition Foundation (CDHNF), is going to be the pediatric component. PEDS-CORI stands for Pediatric Endoscopy Database System–Clinical Outcomes Research Initiative. PEDS-CORI was established in 1999 and initiated at two sites in 2000. The Oregon center is being used as a repository, but a unique component of the Pediatric CORI database is inclusion of biopsy and histologic data. It is an incredibly powerful tool that can be used for prospective trials and for retrospective studies looking at how endoscopic practices are done. Q: How about focusing on high-risk patients? Have we considered a registry for patients with Barrett's esophagus or those with severe esophagitis who need surgical management? To follow all the refluxers who come to us may be a problem. A: CORI can do that. The plan with CORI is essentially to be all-inclusive. We hope that every single pediatric gastroenterologist who does endoscopy will include his or her center in the CORI database. We can do it a lot quicker than adult gastroenterologists can. Essentially the software is a report generator that allows you to consolidate into one report the patient's demographic data, the indications for the procedure, the actual findings during the procedure, and the images that you take. The report can be used in clinical practice, to send to your referring physician as part of the medical record, but the report also goes to a national repository, where it can be used. A: Let's say you have an idea for studying Barrett's esophagus. You go to the National Institutes of Health (NIH) or to industry and propose a study. You obtain the resources, and then you go to CORI. The study is approved by CORI, and you become the investigator who does that study. For a specific hypothesis-driven or question-driven project, a separate protocol will be generated and reviewed by the steering committee. If approved, the proposal will get an IRB review and informed consent for that. Q: If the idea is to follow up a patient from birth to adulthood, then we need to have objective manifestations of disease, especially in the first few months or years of life. Are there signs or symptoms that can be detected by the parents? Somebody mentioned the word heartburn. There's no way we would know when a young child is experiencing heartburn because heartburn is an eminently subjective manifestation. According to some investigators, you can detect heartburn only when a child is 6 to 8 years of age. Furthermore, in adults or even in adolescents, differentiating chronic heartburn from episodic heartburn is important. The two are very different conditions and may have different outcomes. A: You're pointing out the importance of how to phrase the question. The question becomes the critical issue. In the infant, the question has to be focused on vomiting. In the adult, the question has to be focused on heartburn. But what is the question? It's not simply, “Do you have heartburn” or “Do you vomit?” I think that takes a meeting in itself to come up with two questions that everybody in this room would accept as going to yield useful information. The question must come first, the cohort group second. A: There are important natural history questions, from both pediatric and adult standpoints, that should be answered. It is up to investigators to begin to develop the infrastructure that will permit these questions to be answered. The role of acid suppression in children is presented in Table 2. In which areas are studies of validation of acid suppression needed? Is research needed to validate the use of acid suppression?TABLE 2: What is the role of acid suppression in children?Although additional information is needed about pharmacology, appropriate dosing, and drug selection, we feel that there are sufficient data to validate the use of acid suppression as stress ulcer prophylaxis in the critically ill or intubated patient. In addition, there are valid data to support the use of acid suppression in patients with duodenal ulcer and gastric ulcer. For the patient at high risk for symptomatic GERD, there are valid data to justify acid-suppressive therapy. These patients include those with erosive esophagitis and congenital dysmotility syndromes, such as repaired tracheoesophageal fistula. In addition, there are data to justify use of effective acid suppression in children with neurologic impairment, cerebral palsy, progressive neurologic disorders, or genetic disorders that are associated with severe mental-motor retardation. Although the data are not as compelling, they suggest that a pathophysiologic link exists with asthma and that asthmatic patients with GERD should be treated with acid suppression. There is a need for additional study in nonerosive gastroesophageal reflux disease and extraesophageal manifestations of GERD. In infants with unexplained irritability, data suggest that most do not have erosive esophagitis and in most situations would not need endoscopy. Conversely, older children with epigastric abdominal pain/heartburn should be evaluated and their stage of disease based on endoscopic findings. The problem with assessing the benefit of acid suppression in children with extraesophageal disorders associated with GERD is that complete acid suppression may be required to control the symptom. Seventy percent suppression may not be sufficient. This patient group includes children with atypical asthma or an upper respiratory disorder and selected patients with an apparent life-threatening event (ALTE). There are no data to suggest that reflux is a major factor in ALTE or sudden infant death syndrome (SIDS). Also included in this atypical group would be patients with cystic fibrosis and aspiration. Q: There are insufficient data to support the use of acid suppression in low–birth-weight infants. In addition, there are potential concerns about increasing sepsis when these infants are given H2 blockers. Acidifying gastric contents may serve a beneficial purpose in these neonates. A: In the pediatric intensive care unit (PICU) there are not yet enough data to demonstrate that use of these drugs in that setting prevents serious sequelae. I think those studies need to be done. Having said that, I am not sure there is any reason to think that the situation would be any different from what prevails in the adult critical care setting. A: In my experience, the use of antisecretory drugs has dramatically reduced the role of the gastroenterologist in critical care units. We very rarely encounter gastrointestinal bleeding as a complication in PICUs any more because most children receive acid-suppressive prophylaxis. A: Because intensivists are able to more effectively control hypovolemia and shock in the premature infant, I think that you cannot conclude a cause-and-effect relationship with acid suppression. Comment: The majority of cystic fibrosis patients have erosive esophagitis, and this population is ideal for studying GER. There is a well-developed database that is an ongoing project. It may be easy to obtain valuable information about chronic GERD. This effort might be the very first step to an investigation of the relationship between reactive airway disease and GER. Q: The importance of Table 2 is the second category, in which conditions are listed for which there currently is no known rationale for antisecretory therapy. In developing studies, is it justifiable to perform an empirical trial, given that there is no proven rationale for antisecretory therapy? In other words, where is the cart, where is the horse? A: I think there are enough pathophysiologic data to justify treatment. But in those patient groups, any study will require total control of acid production. In other words, if the ear, nose, and throat (ENT) specialists believe that one episode of reflux is sufficient to produce hoarseness and laryngitis, effective therapy must be able to prevent even one reflux episode. With our current pharmacological therapy, is it possible to make a patient achlorhydric? We don't know that yet.