Diagnostic Accuracy of Tests in Pediatric Gastroesophageal

Reux Disease
Rachel J. van der Pol, MD1,*, Marije J. Smits, MSc1,*, Leonie Venmans, PhD3, Nicole Boluyt, MD, PhD2,
Marc A. Benninga, MD, PhD1, and Merit M. Tabbers, MD, PhD1
Objective To systematically review the literature evaluating the diagnostic accuracy of commonly used diagnostic
tests over conventional history taking and physical examination in children #18 months and >18 months suspected
of gastroesophageal reflux disease (GERD).
Study design We searched Medline, Embase, and the Cochrane database for studies assessing the diagnostic
accuracy of pH-metry, pH-impedance, esophagogastroscopy, barium contrast study, scintigraphy, and empirical
treatment as diagnostic tools. Quality was assessed according to Quality Assessment of Studies of Diagnostic Accuracy Included in Systematic Reviews criteria.
Results Of the 2178 studies found, 6 studies were included, containing 408 participants (age 1 month-13.6 years)
and 145 controls (age 1 month-16.9 years). Studies included children with GERD symptoms; 1 included an atypical
presentation. In all the studies, the diagnostic accuracy of pH-metry was investigated, and in 2 studies esophagogastroscopy was investigated as well. Sensitivity and specificity were calculated in 3 studies. The range of reported
sensitivity and specificity was broad and unreliable because of poor methodological quality according to Quality Assessment of Studies of Diagnostic Accuracy Included in Systematic Reviews criteria and inadequate study design.
Conclusion Diagnostic accuracy of tests in children suspected of GERD remains unclear and implications for
practice are hard to give. There is an urgent need of well-designed randomized controlled trials where the effect
of treatment according to specific signs and symptoms will be compared with the effect of treatment based on
the results of additional diagnostic tests, for patient relevant outcomes. (J Pediatr 2013;162:983-7).

astroesophageal reflux (GER) is a physiologic process. Regurgitation occurs in over 70% of infants multiple times
a day, but it tends to disappear by the age of 12-14 months.1,2 Gastroesophageal reflux disease (GERD) is defined
and diagnosed when GER leads to troublesome symptoms and/or complications.3 In 2009, GERD prevalence was estimated to be 12.3% in North American infants and 1% in older children.4 Troublesome symptoms in infants may include
excessive crying, back arching, regurgitation, and irritation around feedings; these could be regarded as nonspecific. In children
and mainly in adolescents, heartburn is the more specific symptom occurring in GERD. Though complaints are often mild, they
are troublesome and may have a significant impact on the wellbeing of the child and family life. Moreover, complications as
esophagitis and hematemesis, failure to thrive, or apparent life threatening events (ALTE) have to be prevented whenever
possible.3,5,6
Diagnosing GERD in pediatric patients is difficult because no gold standard exists, and not one combination of symptoms is
conclusive. Currently, the diagnosis of GERD is based on history and physical examination. This approach might be considered
as the gold standard. However, there is a need to quantify GERD in a more objective way because the GERD diagnosis is
subject to free interpretation and is probably overdiagnosed.7 It may mimic disorders such as cows milk allergy and eosinophilic esophagitis.8,9
Tests for GERD can be divided into 2 categories: tests with the ability to measure reflux events (pH-metry, pH-impedance,
barium contrast studies, and scintigraphy) and tests to detect the consequences of reflux events (esophagogastroscopy).
The most widespread test used to quantify GERD is 24-hour pH-metry. A pH < 4 in the esophagus is generally considered as an
acid reflux episode.3 Acid exposure is expressed as the reflux index (RI, % of time a pH < 4 was measured), for which currently no
evidence based pediatric normal values exist. The European Society for Pediatric Gastroenterology, Hepatology, and Nutrition
and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines consider a RI >7%
as abnormal, a RI <3% as normal, and between 3% and 7% as indeterminate.

ALTE
GER
GERD
PPI
QUADAS
RI
SR

Apparent life threatening event

Gastroesophageal reflux
Gastroesophageal reflux disease
Proton pump inhibitor
Quality Assessment of Studies of Diagnostic Accuracy Included in Systematic Reviews
Reflux index
Systematic reviews

From the Departments of 1Pediatric Gastroenterology

and Nutrition and 2Pediatrics, Emma Childrens Hospital,
Academic Medical Center, Amsterdam, The
Netherlands; and the 3Pediatric Association of The
Netherlands, Utrecht, The Netherlands
*Contributed equally.
The authors declare no conflicts of interest.
0022-3476/$ - see front matter. Copyright 2013 Mosby Inc.
All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2012.10.041

983

THE JOURNAL OF PEDIATRICS

www.jpeds.com

In infants, however, frequency of feeds and buffering

capacities of milk may confound outcomes of pH-metry studies.10 In addition to 24-hour pH-metry, 24-hour
pH-impedance measurement (pH-impedance) was developed.11 Equipped with a pH sensor and multiple electrode
pairs, it measures the conductivity of liquid, gas, or mixed
contents in the esophagus and is able to detect non-acid
and alkaline reflux besides acid reflux.12,13
Barium contrast studies consist of a series of radiographs
of the esophagus and stomach using a barium emulsion to
track swallows and possible reflux, which sometimes reveal
structural anatomic causes underlying GERD.14 In gastroesophageal scintigraphy, patients consume a 99technetium labeled meal prior to start of the scans, and postprandial reflux
becomes visible when labeled stomach contents move upwards in the esophagus.3
Reflux esophagitis, one of the complications of GERD, can
be measured by esophagogastroscopy. This enables both macroscopic and microscopic grading of the esophageal wall. To
date, there are insufficient data to support the use of histology
in diagnosing GERD. Currently, the main reason for taking biopsies is to exclude other diseases causing esophagitis such as
eosinophilic esophagitis, Crohns disease, and infections.3,15
Finally, a trial with an antireflux agent may be used to diagnose GERD. A proton pump inhibitor (PPI) is often the
agent of choice and an empiric trial of 2-4 weeks is common.3
Data on sensitivity and specificity is scarce, both in adults and
children, amplified by the fact that GERD symptoms may improve spontaneously or respond by a placebo effect.3
The accuracy of the above-mentioned tests is unclear, and,
therefore, it is questionable if these more invasive and expensive tools should be used. We carried out a systematic review
to evaluate the accuracy of pH-metry, pH-impedance of
esophagogastroscopy, barium contrast study, scintigraphy,
and diagnostic treatment compared with conventional
history and physical examination when diagnosing GERD.

Methods
A clinical librarian searched Medline, Embase, and the
Cochrane Database of systematic reviews (SR) electronic database for SRs, and clinical studies from inception to May
2012. The key words used to describe the study population
were esophagogastroscopy, pH-metry, pH-impedance,
gastric emptying scintigraphy, barium radiography,
GER, GERD, heartburn, extraesophageal symptoms,
reflux esophagitis, infant, child, and adolescent
(medical subject headings and all fields). No language restriction was applied. Reference lists of reviews and included
studies were searched for additional studies. The full search
strategy is available from the authors.
Two reviewers independently selected the abstracts of identified studies for suitability. Inclusion criteria were: (1) the
study was an SR or clinical study; (2) children were aged 018 years presenting with signs and symptoms (through history
or clinical examination) suggestive of GERD; (3) the aim of the
study was to evaluate the diagnostic accuracy of esophagogas984

Vol. 162, No. 5

troscopy, pH-metry, pH-impedance (symptom index, symptom sensitivity index, and symptom association probability
had to be given), scintigraphy, barium swallow/radiograph of
esophagus/stomach, or diagnostic treatment (at least 1 week
of treatment compared with history and physical examination); and (4) the study had to use a control group. Exclusion
criteria were: (1) no definition of GERD; (2) patients who had
a disease frequently related to GERD (eosinophilic esophagitis,
malformation of the esophagus, [congenital] hernia diaphragmatica, achalasia, cystic fibrosis, gastric paresis, systemic sclerosis, children with neurologic impairment, cows milk
allergy, and rumination syndrome); (3) patients who had
undergone surgical therapy; and (4) children who were treated
for GERD during history and physical examination or prior to
the investigated diagnostic test or vice versa (and the GERD
therapy was not the investigated diagnostic tool).
All potentially relevant studies and the studies for which
the abstracts did not provide sufficient information for inor exclusion were retrieved as full articles.
Two reviewers assessed methodological quality of all identified studies by the Quality Assessment of Studies of Diagnostic Accuracy Included in Systematic Reviews (QUADAS)
checklist.16 Because the revised QUADAS checklist was published after the quality assessment process took place, we
were not able to incorporate this newer checklist.17 The revised QUADAS checklist offers additional and improved features and has improved in distinguishing between bias and
applicability and is capable of rating risk of bias. From the
original standardized list, we choose 11 items (scored yes,
no, or unclear) that could optimally differentiate for methodological quality (Table I; available at www.jpeds.com).
Calculations on summary scores are not provided because
they ignore the importance of individual items and because
cut-off values on what is a good or bad score will be
arbitrarily determined; these results may be misleading.16 In
general, the more items answered with yes, the higher
methodological quality is presumed.
Structured data extraction was performed by 2 reviewers
independently. Data derived from included articles contained items such as author and year of enrollment, diagnostic method, study setting, methods, patient characteristics,
number of participants and controls, index test and execution, sensitivity, and specificity. Because symptom presentation and pathophysiology is different in infants (#18
months) and children, we choose to extract data, if possible,
for infants and children separately. If disagreement between
the 2 reviewers existed, consensus was formed, or a third reviewer (M.T.) made the final judgment.

Results
The search generated 2178 studies, of which 106 met our
inclusion criteria (Figure; available at www.jpeds.com). No
valid SR was encountered. After retrieving the full-text
articles, 100 articles were excluded because of the lack of
a control group, comparison between 2 diagnostic tests,
and, therefore, no comparison with history and physical
van der Pol et al

ORIGINAL ARTICLES

May 2013
examination, use of antireflux medication during the study,
and/or 1 of the other inclusion and/or exclusion criteria
(Figure).
The 6 studies included were all conducted in tertiary centers from Europe (Italy, Spain, and Belgium). Data from 408
participants (age 1 month-13.6 years) and 145 controls (age 1
month-13.6 years) were included. Because of the heterogeneity, especially for symptom presentation of GERD between
the included studies with regard to all participants, diagnostic, and outcome measures, a meta-analysis was not possible,
and all studies are, therefore, discussed separately. Study
characteristics are presented in Table II (available at www.
jpeds.com).
Five studies included children referred for symptoms and
signs of GERD that could be regarded as troublesome (eg, regurgitation, vomiting, hematemesis, weight failure, and recurrent pneumonia),18-22 and 1 study included children
with ALTE.23 The latter study was included because ALTE
was regarded as a possible presentation of GERD. This study
found no relationship between pH drops and the onset of apneas, the latter being the possible cause of ALTE.
All 6 included studies assessed the diagnostic value of pHmetry (5 by glass electrode and 1 antimone electrode),22 and
2 studies assessed the diagnostic value of esophagogastroscopy (macroscopy and histology) as well.21,22 Cut-off values
for pH-metry were defined differently as 1 or 2 SDs of the
mean of the control group in 4 studies.18-21 In 1 study, no
cut-off values were defined.23 In the last study, cut-off values
were defined without further reference to literature or explanation.22 In 2 studies assessing esophagogastroscopy, macroscopic criteria were defined differently. Histology was graded
in both studies equally as described in Table II.
In 3 studies, it was possible to calculate sensitivity for pHmetry of the extracted data (Table III; available at www.jpeds.
com).19,20,22 Sensitivity ranged from 41%-81%. Of the 2
studies performing esophagogastroscopy, sensitivity was
calculated for macroscopy from the extracted data.21,22
Both sensitivity and specificity could be calculated for
histology.21,22 Hence, no studies were retrieved fulfilling
our inclusion criteria assessing impedance, scintigraphy,
barium swallow/radiograph of esophagus/stomach, and/or
a diagnostic treatment.
Results of the methodological quality are presented in
Table I. In all included studies, the patient groups were
representative for those patients who would receive
a diagnostic test in clinical practice. Furthermore, selection
criteria were clearly described, and execution of the index
test was described in sufficient detail to permit replication.
Because GERD signs and symptoms are not distinctive and,
therefore, difficult to diagnose, it was unclear if the reference
standard (signs and symptoms) used in the 6 included
studies, was correctly classifying the target condition.
However, in all studies, except 1,18 signs and symptoms
were clearly described. Only in the study assessing ALTE and
GER, the time period between presentation of the signs and/
or symptoms (ALTE) and time of measurement was
reported, which was 24-48 hours.23 In the 2 studies assessing

histology, assessors were blinded, but blinding procedure for

macroscopy was unclear.21,22 For pH-metry, outcome
assessors were blinded for the clinical profile of the
included patients in 3 studies. Also, outcome assessors were
uninformed about the pH-metry results when interpreting
the results of history taking and physical examination.20,21,23
In 2 studies results were assessed unblinded,18,19 and in 1
study this was unclear.22 Neither uninterruptible nor
intermediate test results were reported; an explanation of
withdrawals was not reported. Surprisingly, withdrawals
were not described in any of the included studies.

Discussion
This systematic review clearly shows that, despite a large
number of publications, there is a lack of high quality studies
of the diagnostic accuracy compared with the current definition of GERD, which is based on history and physical examination. Therefore, the accuracy of tests in children #18
months and >18 months suspected of GERD remains unclear.
Nearly all studies investigating the accuracy of pH-metry
used glass electrode catheters, 18-21,23 but ion sensitive field
effect transistor catheters are preferred nowadays because
of the most accurate in vivo measurements of acid exposure
time.24 In the study that included children with ALTE, the
absence of a relationship between apneas and pH drops
might be due to the fact that pH-metry can only detect
acid reflux.25 Apneas might not be triggered by acid GER
but could be due to nonacid reflux or reasons other than
GERD. Considering the low overall amount of detected pH
drops below 4 (116 in total in 18 children, measured during
1 night), the question arises whether this population suffered
from GERD at all. Furthermore, literature is inconsistent regarding the association of GER and ALTE,26 and pathologic
central and obstructive apnea.27,28 Up until now, it is not
convincingly shown that these are related.29
Although pH-impedance is increasingly popular,30,31 our
search did not retrieve any suitable articles on the accuracy
of pH-impedance in children. The main reason is that
pH-impedance is frequently studied with other tests and,
therefore, not compared with history and physical examination, which is 1 of our inclusion criteria. pH-impedance is
currently the only tool assessing acid, weakly acid, and alkaline reflux, proximal extent and nature of the reflux episodes
being gas, liquid, or mixed. These types of GER are thought
to play an important role in childhood GERD, and it appears that using pH-impedance adds significant benefit in
correlating symptoms and GER in infants.32 The European
Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines recommend,
partly based on expert opinion, to consider pH-impedance
in persisting irritability in infants, despite conservative measures, in children older than 18 months with regurgitation
and vomiting, or in children with ALTE with possible symptoms and signs of GERD.33 The European Medical Agency

Diagnostic Accuracy of Tests in Pediatric Gastroesophageal Reflux Disease

985

THE JOURNAL OF PEDIATRICS

www.jpeds.com

recommends to perform pH-impedance in therapeutic studies to truly quantify and monitor symptom association results.34 However, whether pH-impedance can be
considered as a diagnostic tool with a proven diagnostic accuracy over history and physical examination has never been
properly investigated.
In studies investigating the value of esophagogastroscopy
for GERD, macroscopic criteria were defined differently,
but histology was equally graded in both studies. Indeed, endoscopy can effectively pinpoint the existence of macroscopic
esophagitis, which is a possible but rare complication of
GERD in younger children. It is still under debate, however,
when to use this invasive technique in children because the
existence of esophagitis correlates poorly with symptoms, especially in infants.21,22 For histology, both included studies
showed a relative high sensitivity and specificity in diagnosing
GERD, and multiple studies showed inconsistent correlation
between histology and GERD symptoms.35 No evidence exists
why these 2 poorly correlate; 1 of the many explanations is
the patchy distribution of lesions of reflux esophagitis.
It could be argued that questionnaires, developed to measure pediatric GERD, could be a better diagnostic test to
quantify GERD. However, the best-validated questionnaire,
the Infant-GER Questionnaire Revised, has a high sensitivity
but through its low specificity, it is still advised to use additional invasive testing.6 In this study, no attempt was made to
distinguish between symptomatic infants with and without
GERD; therefore, it is difficult to implement this tool in clinical practice. In clinical practice, diagnosis of GERD is based
on the presence of bothersome symptoms and/or complications. However, because this is the only tool quantifying
symptoms of infant GERD, it is a useful instrument to measure change in symptoms, for example, in research settings.
Reported outcomes in the included studies cannot guide
clinical practice. Although many reports exist giving guidance on which test to perform in clinical practice,32,36-38
the evidence based value of diagnostic tools remains unclear,
and the question whether these more invasive and expensive
tools should be used remains unanswered.
In adults, it is recommended to perform esophagogastroscopy when troublesome dysphagia is present or when an
empiric trial of PPIs failed to diminish complaints after 1
or 2 months. When no esophagitis is present, pH-metry
and pH-impedance should be considered.39 This is especially
of value in patients with atypical GERD symptoms, other
than the classic heartburn and regurgitation.31 For adults,
normal values for pH-metry as well as for pH-impedance
are firmly established. Moreover, enhanced intercellular
space dilatation on histology can be an indication of nonerosive (nonacidic) reflux disease and may, therefore, be useful
in diagnosing GERD refractory to PPI therapy in adults.39
However, even in adults, this is still fully under investigation,
and future studies will elaborate if medical agents interfering
with dilated intracellular spaces need to be developed. Extrapolating adult data toward children and infants might
confound for multiple reasons, mainly because of the difference in GERD symptoms in younger children.
986

Vol. 162, No. 5

It could be hypothesized that if GERD symptoms are present in an infant or child, but no warning signals like hematemesis or failure to thrive exist, further diagnostic testing
is unnecessary. A thorough work-up is warranted when differential diagnosis of signs and symptoms of GERD contains
any cause other than GERD that requires further examination. The test to be used should depend on each specific situation. For example, when food allergies or eosinophilic
esophagitis are suspected, these should be excluded before assuming GERD is present. Especially with the latter, a condition with an increasing incidence worldwide,9,40 symptom
presentation is not distinguishable from classic GERD in infants.41 In older children, dysphagia is often present. However, diagnosis can only be made through endoscopy.
Alternatively, the decision to perform further diagnostic
work-up and the choice of the diagnostic tool used could
also be approached from a therapeutic perspective. Current
therapy of GERD consists of conservative measures and
acid reducing agents. In common practice, these agents are
frequently prescribed without a prior pH-metry. Therefore,
the therapeutic consequences of additional tests will, in
most cases, be very limited. Even if reflux esophagitis is found
during esophagogastroscopy, GERD therapy will still focus
on acid suppression. Only if signs and symptoms persist during acid suppression or, as previously mentioned, a diagnosis
other than GERD is considered, additional diagnostic testing
could be considered.
Research of high quality in this patient population is challenging, amplified by the difficulties of recruiting proper research groups and defining clear selection criteria. Apart
from ethical dilemmas of putting children through possibly
unnecessary invasive procedures, parents are not likely to
give permission to subject their child to such research.42
Most patients eligible for such studies are often missed because
most of them take place in tertiary centers, whereas the majority of GERD related problems are seen in primary care. n
The authors would like to thank Arnold Leenders for all his effort in
conducting the search strategy.
Submitted for publication Jun 25, 2012; last revision received Aug 23, 2012;
accepted Oct 16, 2012.
Reprint requests: Rachel J. van der Pol, MD, Department of Pediatric
Gastroenterology and Nutrition, Emma Childrens Hospital, Academic Medical
Center, Meibergdreef 9, C2-312, 1105 AZ Amsterdam, The Netherlands. Email: r.j.vanderpol@amc.nl

References
1. Nelson SP, Chen EH, Syniar GM, Christoffel KK. One-year follow-up of
symptoms of gastroesophageal reflux during infancy. Pediatrics 1998;
102:E67.
2. Nelson SP, Chen EH, Syniar GM, Christoffel KK. Prevalence of symptoms of gastroesophageal reflux during childhood: a pediatric
practice-based survey. Pediatric Practice Research Group. Arch Pediatr
Adolesc Med 2000;154:150-4.
3. Vandenplas Y, Rudolph CD, Di Lorenzo C, Hassall E, Liptak G, Mazur L,
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

van der Pol et al

ORIGINAL ARTICLES

May 2013

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.
14.
15.

16.

17.

18.
19.

20.

21.

22.

23.

Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN). J Pediatr Gastroenterol Nutr 2009;49:498-547.
Kothari S, Nelson SP, Wu EQ, Beaulieu N, McHale JM, Dabbous OH.
Healthcare costs of GERD and acid-related conditions in pediatric patients, with comparison between histamine-2 receptor antagonists and
proton pump inhibitors. Curr Medical Res Opin 2009;25:2703-9.
Gold BD. Is gastroesophageal reflux disease really a life-long disease: do
babies who regurgitate grow up to be adults with GERD complications?
Am J Gastroenterol 2006;101:641-4.
Sherman PM, Hassall E, Fagundes-Neto U, Gold BD, Kato S, Koletzko S,
et al. A global, evidence-based consensus on the definition of gastroesophageal reflux disease in the pediatric population. Am J Gastroenterol
2009;104:1278-95.
Hassall E. Over-prescription of acid-suppressing medications in infants:
how it came about, why its wrong, and what to do about it. J Pediatr
2011;160:193-8.
Orenstein SR. Symptoms and reflux in infants: Infant Gastroesophageal
Reflux Questionnaire Revised (I-GERQ-R)utility for symptom tracking
and diagnosis. Curr Gastroenterol Rep 2010;12:431-6.
Liacouras CA, Furuta GT, Hirano I, Atkins D, Attwood SE, Bonis PA,
et al. Eosinophilic esophagitis: updated consensus recommendations
for children and adults. J All Clin Immunol 2011;128:3-20.
Mitchell DJ, McClure BG, Tubman TR. Simultaneous monitoring of
gastric and oesophageal pH reveals limitations of conventional oesophageal pH monitoring in milk fed infants. Arch Dis Child 2001;84:273-6.
Skopnik H, Silny J, Heiber O, Schulz J, Rau G, Heimann G. Gastroesophageal reflux in infants: evaluation of a new intraluminal impedance technique. J Pediatr Gastroenterol Nutr 1996;23:591-8.
Peter CS, Wiechers C, Bohnhorst B, Silny J, Poets CF. Detection of small
bolus volumes using multiple intraluminal impedance in preterm infants. J Pediatr Gastroenterol Nutr 2003;36:381-4.
Wenzl TG. Investigating esophageal reflux with the intraluminal impedance technique. J Pediatr Gastroenterol Nutr 2002;34:261-8.
Simanovsky N, Buonomo C, Nurko S. The infant with chronic vomiting:
the value of the upper GI series. Pediatr Radiol 2002;32:549-50.
Dahms BB. Reflux esophagitis: sequelae and differential diagnosis in infants and children including eosinophilic esophagitis. Pediatr Dev Pathol
2004;7:5-16.
Whiting P, Rutjes AWS, Reitsma JB, Bossuyt PMM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of
diagnostic accuracy included in systematic reviews. BMC Med Res
Methodol 2003;3:25.
Whiting PF, Rutjes AWS, Westwood ME, Mallett S, Deeks JJ, Reitsma JB,
et al. QUADAS-2: a revised tool for the quality assessment of diagnostic
accuracy studies. Ann Intern Med 2011;155:529-36.
Boix-Ochoa J, Lafuente J, Gil-Vernet J. Twenty-four-hour esophageal
pH monitoring in gastroesophageal reflux. J Pediatr Surg 1980;15:74-8.
Da Dalt L, Mazzoleni S, Montini G, Donzelli F, Zacchello F. Diagnostic
accuracy of pH monitoring in gastro-oesophageal reflux. Arch Dis Child
1989;64:1421-6.
Cucchiara S, Staiano A, Gobio Casali L, Boccieri A, Paone FM. Value of
the 24 hour intraoesophageal pH monitoring in children. Gut 1990;31:
129-33.
Cucchiara S, Minella R, DArmiento F, Franco M, Iervolino C,
Campanozzi A, et al. Histologic grading of reflux esophagitis and its relationship with intra-oesophageal and intragastric pH variables. Eur J
Gastroenterol Hepatol 1993;5:621-6.
Ravelli AM, Villanacci V, Ruzzenenti N, Grigolato P, Tobanelli P,
Klersy C, et al. Dilated intercellular spaces: a major morphological feature of esophagitis. J Pediatr Gastroenterol Nutr 2006;42:510-5.
Kahn A, Rebuffat E, Sottiaux M, Blum D, Yasik E. Sleep apneas and acid
esophageal reflux in control infants and in infants with an apparent lifethreatening event. Biol Neonate 1990;57:144-9.

24. Hemmink GJM, Weusten BLAM, Oors J, Bredenoord AJ, Timmer R,

Smout AJPM. Ambulatory oesophageal pH monitoring: a comparison
between antimony, ISFET, and glass pH electrodes. Eur J Gastroenterol
Hep 2010;22:572-7.
25. Tolia V, Vandenplas Y. Systematic review: the extra-oesophageal symptoms of gastro-oesophageal reflux disease in children. Alimen Pharm
Ther 2009;29:258-72.
26. Vandenplas Y, Hauser B. Gastro-oesophageal reflux, sleep pattern, apparent life threatening event and sudden infant death. The point of
view of a gastro-enterologist. Eur J Pediatr 2000;159:726-9.
27. Wenzl TG, Schenke S, Peschgens T, Silny J, Heimann G, Skopnik H. Association of apnea and nonacid gastroesophageal reflux in infants: investigations with the intraluminal impedance technique. Pediatr Pulmonol
2001;31:144-9.
28. Peter CS, Sprodowski N, Bohnhorst B, Silny J, Poets CF. Gastroesophageal reflux and apnea of prematurity: no temporal relationship. Pediatrics 2002;109:8-11.
29. Di Fiore JM, Arko M, Whitehouse M, Kimball A, Martin RJ. Apnea is not
prolonged by acid gastroesophageal reflux in preterm infants. Pediatrics
2005;116:1059-63.
30. Rohof WOA, Hirsch DP, Boeckxstaens GEE. Pathophysiology and management of gastroesophageal reflux disease. Min Gastroenterol Dietol
2009;55:289-300.
31. Kahrilas PJ, Shaheen NJ, Vaezi MF, Hiltz SW, Black E, Modlin IM, et al.
American Gastroenterological Association Medical Position Statement
on the management of gastroesophageal reflux disease. Gastroenterology
2008;135:1383-91.
32. Mousa HM, Rosen R, Woodley FW, Orsi M, Armas D, Faure C, et al.
Esophageal impedance monitoring for gastroesophageal reflux. J Pediatr
Gastroenterol Nutr 2011;52:129-39.
33. Loots CM, Benninga MA, Davidson GP, Omari TI. Addition of pHimpedance monitoring to standard pH monitoring increases the yield
of symptom association analysis in infants and children with gastroesophageal reflux. J Pediatr 2009;154:248-52.
34. European Medicines Agency (EMA)[homepage on the internet]. Guideline on the evaluation of drugs for the treatment of Gastro-oesophageal
reflux disease-2011 Available from: http://www.ema.europa.eu/docs/en_
GB/document_library/Scientific_guideline/2010/01/WC500052741.pdf.
Accessed Jan 20, 2012.
35. Gupta SK, Hassall E, Chiu Y-L, Amer F, Heyman MB. Presenting symptoms of nonerosive and erosive esophagitis in pediatric patients. Dig Dis
Sci 2006;51:858-63.
36. Fike F, Mortellaro V, Pettiford J, Ostlie D, St Peter S. Diagnosis of
gastroesophageal reflux disease in infants. Pediatr Surg Int 2011;27:
791-7.
37. Rosen R, Hart K, Nurko S. Does reflux monitoring with multichannel
intraluminal impedance change clinical decision making? J Pediatr Gastroenterol Nutr 2011;52:404-7.
38. Loots CM, Benninga MA, Omari TI. Gastroesophageal reflux in pediatrics: (patho)physiology and new insights in diagnostics and treatment.
Min Pediatr 2012;64:101-19.
39. Weber C, Davis CS, Fisichella PM. Current applications of evolving
methodologies in gastroesophageal reflux disease testing. Dig Liver Dis
2011;43:353-7.
40. Cherian S, Smith NM, Forbes DA. Rapidly increasing prevalence of eosinophilic oesophagitis in Western Australia. Arch Dis Child 2006;91:
1000-4.
41. Straumann A, Aceves SS, Blanchard C, Collins MH, Furuta GT, Hirano I,
et al. Pediatric and adult eosinophilic esophagitis: similarities and differences. Allergy 2012;67:477-90.
42. Henschel AD, Rothenberger LG, Boos J. Randomized clinical trials in
childrenethical and methodological issues. Curr Pharm Des 2010;16:
2407-15.

Diagnostic Accuracy of Tests in Pediatric Gastroesophageal Reflux Disease

987

THE JOURNAL OF PEDIATRICS

www.jpeds.com

Vol. 162, No. 5

Table I. Summary of methodological quality of included studies according to the QUADAS checklist

Was the spectrum of patients representative of the patients who would

receive the test in practice?
Were selection criteria clearly described?
Is the reference standard likely to correctly classify the target condition?
Is the time period between reference standard and index test short enough
to be reasonably sure that the target condition did not change between
the 2 tests?
Was the execution of the index test described in sufficient detail to permit
replication of the test?
Was the execution of the reference standard described in sufficient detail to
permit its replication?
Were the index test results interpreted without knowledge of the results of
the reference standard?
Were the reference standard results interpreted without knowledge of the
results of the index test?
Were the same clinical data available when test results were interpreted as
would be available when the test is used in practice?
Were not interpretable/intermediate test results reported?
Were withdrawals from the study explained?

Boix
1980

Da Dalt
1989

Cucchiara
1990

Kahn
1990

Cucchiara
1993

Ravelli
2006

1
0
0

1
0
0

1
0
0

1
0
1

1
0
0

1
0
0

0/0/1*

1/0/1*

0/0/1*

1/0/1*

2
2

2
2

2
2

2
2

2
2

2
2

1, yes; 2, no; 0, unclear.

*pH-metry/histology/macroscopy.

Figure. Flowchart: result of search strategy.

987.e1

van der Pol et al

May 2013

Diagnostic Accuracy of Tests in Pediatric Gastroesophageal Reflux Disease

Table II. Study characteristics

Study

Diagnostic method/setting

Number of subjects
(mean/median,
age range)

Boix-Ochoa
198018

24 h pH-metry (glass electrode)

Pediatric gastroenterology
clinic

I: 44
(2-18 mo)
C: 20
(2 mo-3 y)

Da Dalt
198919

24 h pH-metry (glass electrode)

Pediatric gastroenterology
clinic

I: 111
(9.3 mo, 1 m-13.5 y)
C: 14
(12.5 mo, 3-68 mo)

Cucchiara
199020

24 h pH-metry (glass electrode)

Pediatric gastroenterology
clinic

I: 114
(GERD: n = 45, 26.6 mo,
2 mo-10 y)
(GERD and esophagitis:
n = 69, 41.3 mo,
1 mo-12 y)
C: 63
(24.02 mo, 2 mo-12 y)

Kahn
199023

8.5 h pH-metry (glass electrode)

Pediatric gastroenterology
clinic

Cucchiara
199321

24 h pH-metry (glass electrode)

Endoscopy (macroscopy/histology)
Pediatric gastroenterology
clinic

Index (test parameters)

Method of execution
of index test

Cases (reference standard)

Controls

Percentage of time pH < 4 (24 h)

Number of refluxes pH < 4
Number of refluxes >5 min
Duration of longest reflux episode
Score of combined parameters
compared to (per 1 SD) controls
Percentage of time pH < 4
Number of refluxes pH < 4
Number of refluxes >5 min
Duration of longest reflux episode
Mean duration all reflux episodes
Normal acid exposure/clearance
time: <2 SD from control mean.
Aberrant pH-metry; abnormal acid
exposure time, and at least 2
other positive parameters
Percentage of time pH < 4 (24 h)
Number of refluxes pH < 4
Acid clearance time (>5 min)
Normal acid exposure/clearance
time: <2 SD from control mean.

Supine, prone, and

semi-seated position
recorded, 8 h each,
4 awake, 4 asleep.
Positioned by manometry

History of recurrent GER, and/or

failure to thrive, and/or weight
loss and/or feeding difficulties.

No history of GER

Normal feeding and daily

activities were maintained.
Positioned by Strobel formula:
87% of nares-LES distance
and fluoroscopy

Admitted subjects with: vomiting

(n = 69), failure to thrive (n = 29),
feeding difficulties (n = 20),
hematemesis (n = 15), recurrent
pneumonia (n = 14), recurrent
wheezing (n = 12), chronic cough
(n = 11), apnea (n = 11), iron
deficiency anemia (n = 10),
epigastric pain (n = 10), irritability
(n = 7)
History of recurrent vomiting,
regurgitation, and/or
complications: weight failure
(n = 45), hematemesis (n = 17),
respiratory symptoms (n = 19)

No history or symptoms
suggestive of GER

I: 10
(9, 4-25 wk)
C: 10
(7, 4-16 wk)

Number of refluxes pH < 4

Number of refluxes >5 min
Acid duration time (in s)

Positioned 3 cm above the

cardia, checked
radiologically

I: 81
(32.9 mo, 2-141 m)
C: 16
(7 mo, 3-154 mo)

Percentage of time pH < 4 (24 h)

Number of refluxes pH < 4
Acid clearance time (>5 min)
Normal acid exposure/clearance
time: <2 SD from control mean.
Aberrant macroscopy: friability,
granularity, erosions, ulcerations
Aberrant histology: basal zone
hyperplasia, papillar elongation,
increased number of eosinophils
and/or neutrophils, mucosal
erosions

Normal feedings were

maintained, only water as
beverage was allowed
Positioned by Strobel formula:
87% of nares-LES distance
and fluoroscopy

Normal feeding and daily

activities were maintained.
Positioned by Strobel formula:
87% of nares-LES distance
and fluoroscopy

Feeding problems due to

maternal inexperience/
anxiety (n = 5), functional
abdominal pain (n = 4),
previous respiratory
complaints (n = 4),
transient decreased
food intake (n = 3)

(continued )
987.e2

ORIGINAL ARTICLES

Full-term infants that had been

found apneic, pale or cyanotic,
loss of tone and consciousness
and had received vigorous
resuscitation. Diagnostic work-up
did not reveal a cause for the ALTE.
History of recurrent emesis and/or
regurgitation (n = 63),
hematemesis (n = 5), asthma
(n = 4), pneumonia (n = 10),
weight failure (n = 23), apnea
(n = 2), irritability (n = 13),
anorexia (n = 12), dysphagia
(n = 6), chest pain (n = 5),
heartburn (n = 3), epigastric
pain (n = 17).

Absence of typical
symptoms GER; functional
abdominal pain (n = 5),
functional constipation
(n = 9), IBS (n = 11),
feeding problems due
to maternal inexperience/
anxiety (n = 19), history
of apnea (n = 10), upper
respiratory infections
(n = 9)
9 full-term and 1 premature
infant, with no history
related to apnea or ALTE.

Study
Ravelli
200622

Diagnostic method/setting
pH-metry (antimony)
Endoscopy (macroscopy/histology)
Pediatric gastroenterology
clinic

Number of subjects
(mean/median,
age range)
I: 48
(3.95 y, 2 mo-11.9 y)
C: 22
(5.85 y, 1-16.9 y)

Index (test parameters)

pH-metry positioned between

9th and 10th dorsal vertebra
and fluoroscopy
Esophagogastroduodenoscopy:
(under sedation). Standard
scopes were used, outer
diameter depending on size
and age. Biopsies: n = 2-4,
minimal 1.5 mm thickness,
3-5 cm above the Z line.

Cases (reference standard)

History of: vomiting, regurgitation,
epigastric pain, heartburn
and/or dysphagia; n = 26;
behavioral symptoms: crying,
fussiness, back arching at
meals; n = 5; respiratory:
nocturnal cough, wheezing,
recurrent pneumonia, apnea;
n = 17.

Suspected food sensitive

enteropathy (n = 11),
H Pylori infection (n = 6),
IBD (n = 4), NSAID
gastropathy (n = 1)

www.jpeds.com

C, control group; IBD, irritable bowel disease; IBS, irritable bowel syndrome; I, intervention group; LES, lower esophageal sphincter; NSAID, nonsteroidal anti-inflammatory drug.

Controls

Percentage of time pH < 4

Number of refluxes pH < 4
Number of refluxes >5 min
Duration of longest reflux episode
Percentage of time pH < 4: mild
(5%-10%), moderate (10%-20%)
and severe (>20%)
Aberrant macroscopy: erosive
lesions
Aberrant histology: basal zone
hyperplasia, papillar elongation,
mucosal erosions, dilatation
of interpapillary vascular spaces;
increased number of neutrophils,
eosinophils and lymphocytes,
erosions or ulcerations and
granulation tissue.

Method of execution
of index test

THE JOURNAL OF PEDIATRICS

987.e3

Table II. Continued

Vol. 162, No. 5

van der Pol et al

ORIGINAL ARTICLES

May 2013

Table III. Sensitivity and specificity

Study

Method

Boix-Ochoa 198018 pH-metry

Number of Patients with

GERD/number of Patients
without GERD (based on signs
and symptoms at presentation)
44/20

Da Dalt 198919

pH-metry

111/14

Cucchiara 199020

pH-metry

114/63

Kahn 199023

pH-metry

10/10

Cucchiara 199321

pH-metry
endoscopy

81/16

Ravelli 2006

pH-metry
endoscopy

48/22

Sens
Unable to calculate sens or spec.
No P values or cut-off values given. Final
score calculated out of different components
was highest in the disease group.
positive pH-metry: n = 45
Sens = 45/111 = 41%
RI parameter:
22 (14.6% GERD n = 45; 21.7% GERD +
oesophagitis n = 69) normal RI.
Sens = 92/114 * 100 = 81%
Duration >5 min reflux parameter:
34 (29.17% GERD n = 45; 30.34% GERD +
oesophagitis n = 69) normal duration.
Sens = 80/114 *100 = 70%
Unable to calculate sens or spec. No P values
or cut-off values given. ALTE versus control
group no statistical difference in number
reflux episodes, lowest pH value, duration
of longest reflux episode and drops in pH.
pH-metry:
Unable to calculate sens for pH metry. No
cut-off values.
Macroscopy:
61.9% = 44 of patients with microscopic
esophagitis had normal macroscopic
result. Abnormal: 81-44 = 27
Sens = 27/81 *100 = 33%
Histology:
n = 10 patients had normal histologic result.
Abnormal = 81-10 = 71
Sens = 71/81 *100 = 88%
pH-metry:
Percentage of time pH < 4: (15 of 29
patients had a positive RI (5.7 < RI > 36).
Sens = 15/29 * 100 = 52%
Macroscopy:
7 of 48 GERD patients had macroscopic
esophagitis
Sens = 7/48 * 100 = 15%
Histology:
40 of 48 patients had histologic esophagitis
Sens = 40/48 * 100 = 83%

Spec

Unable to calculate spec. Values of

controls used as normal values.
Unable to calculate spec. Values of
controls used as normal values.
Therefore, Spec will always be 100%.

pH-metry:
Unable to calculate spec for pH metry.
No cut-off values.
Macroscopy:
Unable to calculate spec for macroscopy.
No data on controls given.
Histology:
Abnormal histologic changes were
absent in controls.
Spec = 16/16 * 100 = 100%
pH-metry:
No controls underwent pH metry
Macroscopy:
Unable to calculate spec since results of
controls are not given
Histology:
In all controls (n = 22), histology was
normal
Spec = 22/22 * 100 = 100%

Sens, sensitivity; Spec, specificity.

Diagnostic Accuracy of Tests in Pediatric Gastroesophageal Reflux Disease

987.e4