What Is Known
- Knowledge regarding prognosis and factors influencing the clinical course of pediatric gastroesophageal reflux disease enables health care professionals to provide accurate information and patient-tailored treatment.
- No overview of prognostic factors in childhood gastroesophageal reflux disease exists in the literature.
What Is New
- Only 4 articles of great heterogeneity were included, highlighting the lack of high-quality research for gastroesophageal reflux disease outcome in children.
- The percentage of children with a diagnosis of gastroesophageal reflux disease with esophagitis that had persisting symptoms and/or were on antireflux medication at follow-up (12 months to >5 years) ranged from 23% (weekly symptoms) to 68% (antireflux medication), depending on definition used.
Gastroesophageal reflux (GER) is considered normal physiology in infants and resolves spontaneously in most cases by the second year of life. It is defined as GER disease (GERD) when it leads to troublesome symptoms and/or complications (1–4). The study of the epidemiology of pediatric GERD is restricted by the lack of consensus over the basic definition and of a well-validated diagnostic tool (5). Epidemiologic estimates of the prevalence of pediatric GERD are based primarily on patient or parental questionnaire surveys on GERD symptoms, and vary widely according to the age group and population studied, ranging from 2% to 15% worldwide (3,4,6–8).
Commonly held “dogma” by pediatricians is the belief that GER(D) is outgrown after infant or early childhood onset. Recent evidence, however, suggests that GERD in some subjects is a chronic, potentially life-long condition that begins in childhood and several studies suggest that the longer the duration of GERD symptoms, the higher the risk of persisting GERD symptoms and long-term sequelae such as Barrett esophagus and esophageal adenocarcinoma (9–14).
In this systematic literature review, we investigated and summarized the quantity and quality of currently available prospective, longitudinal observational studies on the course of GERD in otherwise healthy children with and without treatment.
Databases of Embase and MEDLINE/PubMed were searched from inception to April 2016. In addition, reference lists of review articles and included studies were searched. No date or language restrictions were applied. The exact reporting guidelines as described in the PRISMA statement (www.prisma-statement.org) were followed. Full methodological information including a summary of the search strategy and study flowchart can be accessed online (Supplemental Digital Content 1, http://links.lww.com/MPG/B68).
The search yielded a total of 5365 references. After removal of duplicates, 3950 references remained and after screening the title and abstract of these references 22 publications were judged potentially relevant and after reading the full-text articles, 4 studies were included (online access only; Supplemental Digital Content, Prisma 2009 Flow Diagram, http://links.lww.com/MPG/B68). No additional studies were identified through bibliographic review of included studies.
Study characteristics are described in Supplemental Digital Content 2, http://links.lww.com/MPG/B69(12,15–17). The study of Orenstein et al (15) consisted of the placebo arm (n = 19 patients) of a randomized, double-blind, placebo-controlled trial evaluating a histamine-2 receptor antagonist and a prokinetic agent, or both. The final analysis in the present study included 10 patients that did not require rescue treatment as mandated by the protocol and completed the 12 months follow-up visit. Two studies were retrospective/prospective cohort studies and did not report baseline GERD symptoms and/or complications (12,16). It is unclear whether the children in these studies received treatment during follow-up; in one study it was reported that 21% of patients (261/1242) used acid-suppressive therapy within 30 days of initial diagnosis (16). In the study by Shepherd et al, a prospective cohort study, all patients received positioning therapy, thickened feeds, antacids and bethanechol, metoclopramide, or cimetidine in case of esophagitis. For the present study, data are only displayed for those patients that had proven esophagitis (macroscopically) at baseline, as no clear symptom-based definition of GERD could be extracted from the article (17).
All studies scored high or moderate risk of bias in at least 2 of the 6 domains (Supplemental Digital Content 2, http://links.lww.com/MPG/B70). Three of the four studies scored high or moderate risk in both the domains study participation and study attrition, indicating selective sampling of participants and high or unclear drop-out rates. There was moderate risk of bias in outcome measurement in 2 studies.
Results of unfavorable outcome with regards to resolution/occurrence of clinical GERD symptoms and endoscopic complications are summarized in Table 1. Three studies assessed the occurrence or persistence of GERD symptoms, related to outcomes of symptom questionnaires and/or to the patient's ongoing need for GERD treatment. Three studies assessed the occurrence of GERD-related complications based upon endoscopy results, of which 1 study used histology to determine esophagitis/Barrett esophagus (15), the other studies used the macroscopic appearance only (12,17). In summary, in the study by Orenstein et al (15), GERD symptoms resolved in all 10 infants without any treatment; however, in all, biopsy findings remained abnormal based upon papillary height and/or basal thickness. In the study by Shepherd et al (17), where 34 of 126 of the included patients had endoscopy-proven esophagitis, 24 of 126 (19.0%) required ongoing medication at 18 months of follow-up; all except 1 patient required surgical intervention. In the study by El-Serag et al, 23% of (18/80) patients had persistent weekly symptoms of heartburn and/or acid regurgitation at >5 years after initial diagnosis. Fourteen patients, of which 21% reported weekly symptoms, underwent follow-up endoscopy, and 3 of 14 patients showed erosive changes (n = 2 mild and n = 1 moderate erosion; no Barrett esophagus). Based upon the study of Ruigomez et al (16) in 1242 GERD patients without a diagnosis of reflux esophagitis at baseline, 18 of 1242 (1.4%) developed a registered diagnoses of esophagitis after mean follow-up of 4 ± 1.9 years; none of the evaluated patients developed a Barrett esophagus.
Prognostic Factors (n = 2 Studies)
Two studies analyzed 7 prognostic factors, using 8 different prognostic determinants in a total of 18 associations with the occurrence of GERD symptoms and/or endoscopic complications at follow-up (Table 2) (12,16). Prescription of proton pump inhibitors with or without histamine receptor antagonists at the time of initial consultation and the initial diagnosis (GERD or heartburn) were associated with a subsequent diagnosis of reflux esophagitis. Sex, age at onset of GERD, and number of visits to primary care physician were not significantly associated.
To our knowledge, no previous reviews on prognosis or prognostic factors of pediatric GERD have been performed. In the present systematic review, only 4 articles of great heterogeneity regarding definitions, populations, outcome measures, and follow-up periods could be included. With regard to these differences, the percentage of children with a diagnosis of GERD with esophagitis that had persisting symptoms and/or were on antireflux medication at follow-up (12 months to >5 years) ranged from 23% (weekly symptoms) to 68% (antireflux medication), depending on definition used (12,15,17). None of the included studies made use of a control group, barring Ruigomez et al, which only included the control group for extraesophageal symptoms, hampering differentiation between persistent childhood onset GERD into adulthood, versus resolution of childhood GERD followed by new onset of adult GERD.
Although GERD-related symptoms resolved in the majority of the infants who did not receive any GERD medication at the end of follow-up in the study by Orenstein et al (15), persisting histologic abnormalities were found in all at 12 months follow-up. This discordant finding of unresolved esophageal inflammation in the face of symptom reduction and/or resolution is in line with other studies, demonstrating that at first presentation, Barrett esophagus patients have little or no “typical” GERD-related symptoms (18,19). In contrast, in the study by El-Serag et al, where 57% (8/14) of patients that had follow-up endoscopy used GERD medication, this was despite the resolution of macroscopic esophagitis in the vast majority of patients, and none showed evidence of Barrett esophagus. Whether these patients were, however, actually having ongoing GERD, nonerosive GERD, or functional dyspepsia could only have been resolved by performing a pH-impedance study, which was not done in any of the studies included in this review. Last, based upon results of the study of Ruigomez et al, that included children in the primary care setting without evidence of esophagitis at baseline, only 1.4% developed a new endoscopy-based diagnosis of esophagitis during minimum follow-up of 5 years and none showed evidence of a Barrett esophagus. No correlation with the presence of GERD symptoms at the time of endoscopy was, however, performed in the present study (16). Based upon a large, cross-sectional, multicenter study of children and adolescents without neurodevelopmental or anatomical abnormalities presenting for upper endoscopy for nonurgent matters, the prevalence of endoscopic erosions and endoscopically suspected Barrett esophagus was 0.02% and 1.41%, respectively. The prevalence of histology-graded esophagitis in this same cohort, on the contrary, was much higher, namely 25.5%, and interestingly these patients reported significantly more GERD symptoms than patients without esophagitis, albeit this association did not remain significant in multivariable analysis (19). It should, however, be noted that the diagnostic yield of endoscopy, both macroscopically as microscopically, in children with symptoms suggesting GERD is not well established (1). Two studies in children undergoing endoscopy due to GERD symptoms respectively found erosive esophagitis in 35% (20) and histologic evidence of esophagitis in 13% of children (21). Whether the cohort of Orenstein et al had a “relapse” or recurrence of symptoms after the time of follow-up, or that a longer follow-up is deemed to evaluate endoscopic healing, however, remains unknown.
A total of 7 potential prognostic factors were studied, and there was a high level of variability with regard to which factors were examined in each study. The majority of evidence suggesting an association between prognostic factors and occurrence of GERD symptoms and/or complications was limited or conflicting. Although earlier age of onset was found to be related with the occurrence of heartburn and/or regurgitation in one study, it was not found to be related with the development of esophagitis in another study (12,16). Unfortunately, neither of the 2 studies assessed prognostic factors for heartburn and/or regurgitation and for esophagitis, hampering direct comparison between study results. Several retrospective adult cohort-based studies have shown evidence for a relationship between an earlier age or longer duration of GERD symptoms and the development of Barrett esophagus (13,22).
The present systematic review highlights the lack of high-quality research for potential prognostic factors and GERD outcome in the pediatric population and shows that further research by means of large follow-up studies on prognosis of childhood GERD and its subsequent serious disease sequelae such as Barrett, erosive esophagitis, strictures, and even adenocarcinoma and factors of influence on prognosis is necessary. For future studies, we recommend using a uniform, world-wide accepted definition for the diagnosis of GERD as by the ESPGHAN/NASPGHAN clinical guidelines. Clearly, before health measures that address infant and childhood GERD can be adopted, additional natural history studies are needed to determine the long-term relevance of this diagnosis. Although factors such as patient age and comorbidities are not modifiable, others factors, such as tobacco exposure, body mass index, and dietary composition, might be (23). If the risk of serious long-term GERD sequelae can be significantly reduced, or even eliminated by intervention in childhood GERD, cost-effective surveillance and/or screening and treatment strategies in “at-risk” pediatric populations could and should be rationally implemented in the near future.
1. Vandenplas Y, Rudolph CD, Di Lorenzo C, et al. Pediatric gastroesophageal reflux clinical practice guidelines: joint recommendations of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN). J Pediatr Gastroenterol Nutr
2. Nelson SP, Chen EH, Syniar GM, et al. One-year follow-up of symptoms of gastroesophageal reflux during infancy. Pediatric Practice Research Group. Pediatrics
3. Nelson SP, Chen EH, Syniar GM, et al. Prevalence of symptoms of gastroesophageal reflux during childhood: a pediatric practice-based survey. Pediatric Practice Research Group. Arch Pediatr Adolesc Med
4. Nelson SP, Kothari S, Wu EQ, et al. Pediatric gastroesophageal reflux disease
and acid-related conditions: trends in incidence of diagnosis and acid suppression therapy. J Med Econ
5. Van der Pol RJ, Smits MJ, Venmans L, et al. Diagnostic accuracy of tests in pediatric gastroesophageal reflux disease
. J Pediatr
6. Costa AJ, Silva GA, Gouveia PA, et al. Prevalence of pathologic gastroesophageal reflux in regurgitant infants. J Pediatr (Rio J)
7. Okimoto E, Ishimura N, Morito Y, et al. Prevalence of gastroesophageal reflux disease
, adults, and elderly in the same community. J Gastroenterol Hepatol
8. Martigne L, Delaage PH, Thomas-Delecourt F, et al. Prevalence and management of gastroesophageal reflux disease
and adolescents: a nationwide cross-sectional observational study. Eur J Pediatr
9. Gold BD. Gastroesophageal reflux disease
: could intervention in childhood reduce the risk of later complications? Am J Med
2004; 117 (suppl 5A):23s–29s.
10. Spechler SJ, Lee E, Ahnen D, et al. Long-term outcome of medical and surgical therapies for gastroesophageal reflux disease
: follow-up of a randomized controlled trial. JAMA
11. Waring JP, Feiler MJ, Hunter JG, et al. Childhood gastroesophageal reflux symptoms in adult patients. J Pediatr Gastroenterol Nutr
12. El-Serag HB, Gilger M, Carter J, et al. Childhood GERD is a risk factor for GERD in adolescents and young adults. Am J Gastroenterol
13. Thrift AP, Kramer JR, Qureshi Z, et al. Age at onset of GERD symptoms predicts risk of Barrett's esophagus. Am J Gastroenterol
14. Balasubramanian G, Singh M, Gupta N, et al. Prevalence and predictors of columnar lined esophagus in gastroesophageal reflux disease
(GERD) patients undergoing upper endoscopy. Am J Gastroenterol
15. Orenstein SR, Shalaby TM, Kelsey SF, et al. Natural history of infant reflux esophagitis: symptoms and morphometric histology during one year without pharmacotherapy. Am J Gastroenterol
16. Ruigomez A, Lundborg P, Johansson S, et al. Follow-up of a cohort of children
and adolescents with gastro-esophageal reflux disease who were free of reflux esophagitis at initial diagnosis. Scand J Gastroenterol
17. Shepherd RW, Wren J, Evans S, et al. Gastroesophageal reflux in children
. Clinical profile, course and outcome with active therapy in 126 cases. Clin Pediatr (Phila)
18. Bonino JA, Sharma P. Barrett esophagus. Curr Opin Gastroenterol
19. Nguyen DM, El-Serag HB, Shub M, et al. Barrett's esophagus in children
and adolescents without neurodevelopmental or tracheoesophageal abnormalities: a prospective study. Gastrointest Endosc
20. El-Serag HB, Bailey NR, Gilger M, et al. Endoscopic manifestations of gastroesophageal reflux disease
in patients between 18 months and 25 years without neurological deficits. Am J Gastroenterol
21. Sheiko MA, Feinstein JA, Capocelli KE, et al. Diagnostic yield of EGD in children
: a retrospective single-center study of 1000 cases. Gastrointest Endosc
22. Sanchis Blanco G, Ibanez Pradas V, Couselo Jerez M, et al. Risk factors for Barrett's oesophagus in children
: a case-control study. Cir Pediatr
23. El-Serag HB, Satia JA, Rabeneck L. Dietary intake and the risk of gastro-oesophageal reflux disease: a cross sectional study in volunteers. Gut