Gastrointestinal Conditions in Children With Autism Spectrum Disorder:

Developing a Research Agenda

Daniel L. Coury, Paul Ashwood, Alessio Fasano, George Fuchs, Maureen Geraghty,
Ajay Kaul, Gary Mawe, Paul Patterson and Nancy E. Jones
Pediatrics 2012;130;S160
DOI: 10.1542/peds.2012-0900N

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pediatrics.aappublications.org/content/130/Supplement_2/S160.full.html

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2012 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014

Gastrointestinal Conditions in Children With Autism
Spectrum Disorder: Developing a Research Agenda
Autism spectrum disorders (ASDs) are a set of complex neuro- AUTHORS: Daniel L. Coury, MD,a Paul Ashwood, PhD,b
developmental disorders defined behaviorally by impaired social in- Alessio Fasano, MD,c George Fuchs, MD,d Maureen
Geraghty, PhD, RD,e Ajay Kaul, MBBS, MD,f Gary Mawe,
teraction, delayed and disordered language, repetitive or stereotypic PhD,g Paul Patterson, PhD,h and Nancy E. Jones, PhDi
behavior, and a restricted range of interests. ASDs represent a significant aDevelopmental/Behavioral Pediatrics, Nationwide Children’s
public health issue with recent estimates indicating that as many as 1% Hospital, Columbus, Ohio; bDepartment of Medical Microbiology
of children in the United States are diagnosed with an ASD.1,2 Many and Immunology, University of California, Davis, Davis, California;
cPediatrics, University of Maryland, School of Medicine,
individuals with ASDs have symptoms of associated medical conditions,
Baltimore, Maryland; dPediatrics/Pediatric Gastroenterology,
including seizures, sleep problems, metabolic conditions, and gastro- University of Arkansas for Medical Sciences, Little Rock,
intestinal (GI) disorders, which have significant health, developmental, Arkansas; eMedical Dietetics, Abbott Laboratories, Columbus,
social, and educational impacts. Gastrointestinal complaints are a com- Ohio; fPediatrics, Cincinnati Children’s Hospital, Cincinnati, Ohio;
gAnatomy and Neurobiology, University of Vermont, Burlington,
monly reported concern for parents and may be related to problem Vermont; hCalifornia Institute of Technology, Pasadena, California;
behaviors and other medical issues such as dysregulated sleep (ATN and iClinical Programs, Autism Speaks, Los Angeles, California
Annual Registry Report, unpublished data, November 2009).3 Despite the KEY WORDS
magnitude of these issues, potential GI problems are not routinely autism, gastrointestinal disease, research
considered in ASD evaluations. This likely reflects several factors, in- ABBREVIATIONS
cluding variability in reported rates of GI disorders, controversies re- ASD—autism spectrum disorder
GI—gastrointestinal
garding the relationship between GI symptoms and the putative causes 5-HT—serotonin
of autism, the limited verbal capacity of many ASD patients, and the lack
Drs Ashwood, Fasano, Fuchs, Geraghty, Kaul, Mawe, and
of recognition by clinicians that certain behavioral manifestations in Patterson contributed equally to this work.
children with ASDs are indicators of GI problems (eg, pain, discomfort, or This manuscript has been read and approved by all authors.
nausea).4–10 This paper is unique and not under consideration by any other
publication and has not been published elsewhere.
Whether GI issues in this population are directly related to the patho-
www.pediatrics.org/cgi/doi/10.1542/peds.2012-0900N
physiology of autism, or are strictly a comorbid condition of ASD remains
to be determined, but clinical practice and research to date indicate the doi:10.1542/peds.2012-0900N

important role of GI conditions in ASDs and their impact on children as Accepted for publication Aug 8, 2012
well as their parents and clinicians.9 Address correspondence to Daniel L. Coury, MD, Professor of
Pediatrics and Psychiatry, The Ohio State University, Chief,
On November 15, 2009, a symposium addressing these issues was or- Developmental & Behavioral Pediatrics, Nationwide Children’s
ganized as an adjunct to the annual meeting of the North American Hospital, 700 Children’s Dr, Timken G-350, Columbus OH
Society for Pediatric Gastroenterology, Hepatology, and Nutrition. A panel 43205-2696

of international experts presented the latest scientific information on PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
pathophysiology, evaluation, and treatment strategies for GI conditions Copyright © 2012 by the American Academy of Pediatrics
in children and adolescents with ASDs. One aim of the meeting was to FINANCIAL DISCLOSURE: The information reviewed in this
article was based on presentations from a symposium and
raise awareness among gastroenterologists and GI researchers of GI
workshop funded by Autism Speaks and cosponsored by the
disorders in the ASD population and to provide clinicians with in- American Academy of Pediatrics GI Subcommittee and the North
formation to improve their clinical practice for these children. The American Society for Pediatric Gastroenterology, Hepatology
symposium addressed 4 major areas of concern for children with ASDs: and Nutrition. The authors have indicated they have no financial
relationships relevant to this article to disclose.
reflux, constipation, diarrhea, and nutrition. Each session reviewed the
state of the evidence, the latest findings on issues such as intestinal
permeability, inflammatory processes, innervation, motility, nutrition,
and the epidemiology, presentation, and clinical management of GI
issues.
The symposium also set the context for a follow-up workshop on No-
vember 16 that focused on identifying and prioritizing the key research
topics for further investigation. The 1-day workshop brought together

S160 COURY et al
Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 SUPPLEMENT ARTICLE

a group of symposium participants and

speakers with a broad range of ex-
pertise in GI and autism clinical prac-
tice, and clinical as well as basic science
research.
This report provides an overview of
findings in GI and ASDs as presented in
the symposium and highlights the key
conclusions from the workshop. Spe-
cifically, the group identified the fol-
lowing 4 topics as priority areas for
further study: epidemiology of GI con-
ditions in ASDs, underlying pathology
(gut-brain connection, immune function,
animal models and genome-microbiome
interaction), treatment and outcome,
and nutrition. A recent consensus report FIGURE 1
Reported prevalence of gastrointestinal disorders in children with ASD.
on GI conditions in ASDs published in
Pediatrics in 20109 put forth 23 rec-
ommendations for the evaluation and
management of GI conditions in ASDs. in children with ASDs and higher rates UNDERLYING BIOLOGY OF GI
Although ASDs are behaviorally defined in all but one study when a control DYSFUNCTION IN ASDs
disorders, current thinking suggests population was used.10
Current State of Knowledge
multiple “autisms” with varying bi- Recommendations The underlying nature of GI dysfunction
ological underpinnings. Some of these
Accurately determining the rates of GI in ASDs and its relationship to etiology
may eventually be identified as having
disorders will require addressing the and ASD symptoms are poorly un-
associated GI symptoms, as has been
significant methodological limitations derstood, and systematic research in
seen with the MET gene. Pending new
of previous studies. These limitations this area has been limited. There is,
evidence on any such relationships, the
include suboptimal or, in some cases, however, emerging evidence relevant to
recommendations in this current arti-
lack of controls, variability in the control ASDs in the areas of immune function,
cle expand upon several of the key
groups used across studies, inclusion the relationship between signaling path-
recommendations made in the con-
of populations of children with ASDs ways of the gut and brain, and genome–GI
sensus statement highlighting areas in
that are heterogeneous, retrospective microbiome interactions. Increasingly,
need of new knowledge.
approaches, bias (selection, referral, evidence supports a combination of
or recall), and/or reliance on parent changes in gut microflora, intestinal
EPIDEMIOLOGY OF GI CONDITIONS report only.22 Developing rigorously permeability, inappropriate immune
IN ASDs designed, prospective population-based response, activation of specific meta-
Several studies have assessed the prev- prevalence studies of defined GI symp- bolic pathways, and behavioral changes
alence and types of GI disorders in chil- toms and disorders in ASD is a high pri- in genetically predisposed individuals.
dren with ASDs. The reported prevalence ority, and a shared data set would enable: Integrating findings across these areas
of any GI disorder in children with ASDs 1. Identification of risk factors includ- into a unifying theory will be critical to
ranges from 9% to 91% (see Fig 1), ab- ing clinical and behavioral indicators understanding the mechanisms and
dominal pain or discomfort ranges from of GI problems manifestations of GI disorders in ASDs.
2% to 41%, constipation from 6% to 45%, 2. Identification of atypical presenta-
diarrhea from 3% to 77%, and persistent Immune Function
tions of GI disorders in ASDs
diarrhea from 8% to 19%3,9,11–21. Al- Many reports have described immune
3. Identification of subpopulations with-
though all the studies have significant abnormalities in ASDs including changes
in ASD that have GI symptoms
methodological limitations, they col- in immune-related gene expression,
lectively indicate unusually high rates 4. Correlation with biomarkers skewed cytokine production, altered
of GI disorders or certain GI symptoms 5. Stratification of treatment groups T-cell function, and enhanced innate

PEDIATRICS Volume 130, Supplement 2, November 2012 S161

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
immune responses.23,24 Mucosal im- affect gut function and sensitivity.35 neophobia, deficits in social interaction
mune cells make up ∼70% of the Recently, studies involving germ-free and verbal and olfactory communica-
immune cells within the body, and mice inoculated with selective bacte- tion, increased stereotyped and re-
dysfunction in these cells may have ria have made clear that commensal petitive motor behaviors, enhanced
adverse consequences for GI function. and probiotic gut microbes can have anxiety, abnormal pain sensitivity and
Endoscopic analyses of children with an influence on brain function and eye blink conditioning, disturbed sleep
ASD and GI symptoms have revealed the behavior.36 patterns, seizures, and deficits in sen-
presence of a subtle, diffuse inflammation Serotonin (5-HT), a highly prevalent sorimotor gating. Some animal models
of the intestinal tract (reviewed in refs signaling molecule in the gut (80%–95% also exhibit neuropathology that is
9 and 25). Characterizing the nature of of 5-HT receptors are found in the gut), frequently seen in autism such as a
this inflammation remains an area in is critical for GI function, and alter- spatially restricted Purkinje cell deficit,
need of further investigation to fully ations in 5-HT signaling are implicated as well as characteristic neurochemi-
understand and to provide further ev- in a number of GI disorders. In both cal changes (5-HT abnormalities) and
idence of its relationship to GI symp- human and animal studies, altered 5- alterations in the immune status in the
toms in individuals with ASDs. HT signaling has been implicated in brain and periphery.46,49 It is critically
Autoimmune responses in children with inflammatory bowel diseases such as important that these animal models of
ASDs and a familial history of autoim- Crohn disease and ulcerative colitis ASD environmental and genetic risk
munity have been reported.25–28 Among and in functional disorders such as factors also be examined for GI symp-
the most common findings in ASD sub- irritable bowel syndrome and chronic toms. It should be noted that treatment
jects and their mothers is an increased constipation.37–42 Recent studies have with potential probiotic bacteria can
presence of autoantibodies directed demonstrated that 5-HT acts as a normalize anxiety-like behavior in colitis
toward central nervous system pro- proinflammatory modulator in the in- and infection mouse models,50 whereas
teins.29–33 In addition, autoantibodies testinal mucosa,43,44 and that circulat- such treatments in normal mice can
that bind to basement membranes ing 5-HT arising from the gut affects induce anxiolytic-like effects in some
of epithelial cells and colocalize with bone growth, with elevated 5-HT levels tests and anxiogenic effects in other
complement proteins are observed in leading to decreased bone density. tests.51–53
the intestinal mucosa of children with Because both GI and bone disorders Intestinal Permeability and Gut
ASD and GI symptoms.34 It has been have been observed in ASD, altered gut Microbiome
speculated that these autoantibodies 5-HT signaling may be associated with
It has been hypothesized that altered
could indicate the presence of inflam- these changes.45 The 5-HT system
intestinal permeability (“leaky gut”)
matory processes and/or an autoim- within the gut may be an important
may play a pathogenic role in autism.
mune component that could affect factor contributing to GI problems in
Indeed, several studies have reported
the integrity of the mucosal barrier children with ASD (see also Nutrition
impaired intestinal barrier function
and contribute to decreased mucosal below).
in ASD.54,55 Moreover, Altieri and col-
barrier integrity; it is also possible
leagues56 recently reported a signifi-
that they indicate previous mucosal
Animal Models cant (P , .01) elevation in the level of
inflammation.
Several environmental factors that in- a bacterial product in the urine of
crease the risk for the development of young autistic children. This molecule,
Gut–Brain Connection and autistic features, including maternal p-cresol, is produced by bacterial strains
Serotonin Signaling in the Gut infection and valproate exposure, have that are found to be elevated in ASD. ASD
The digestive system has its own com- been used successfully in rodents to children have higher counts and more
plex (enteric) nervous system that generate animal models of these ASD species of Clostridia, Bacteroidetes, and
regulates gut functions such as motility environmental risk factors (reviewed Firmicutes than controls.57,58
and mucosal secretion. It is now well in ref 46). There is also a non-human One critical feature of the GI tract is its
accepted that the gut–brain axis is a 2- primate model involving administra- ability to regulate the trafficking of
way street. For example, in addition to tion of antibrain antibodies to pregnant macromolecules between the environ-
traditional autonomic projections that monkeys47 and activating the maternal ment and the host through a barrier
regulate digestive reflexes, signals immune system.48 These environmen- mechanism. Together with gut-associated
traveling from the gut to the brain in- tal and genetic models exhibit vari- lymphoid tissue and the neuroendocrine
fluence satiety, and stress and anxiety ous characteristics of autism including network, the intestinal epithelial barrier,

S162 COURY et al
Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 SUPPLEMENT ARTICLE

with its intercellular tight junctions, availability is altered in ASD children and nutritional factors affecting nu-
controls the trafficking of macro- with GI symptoms versus age-matched trition.
molecules.59 The protein zonulin is controls with GI symptoms. Another Nutritional status and nutrient intake
a component of intercellular tight junc- critical need is the determination of are inextricably related in children with
tions that is involved in regulating gut unique biomarkers (eg, cytokines, glu- autism. The assessment of nutritional
permeability.60,61 Small-intestinal expo- tathione redox, antibodies, and zonulin) status in ASDs has been recently dis-
sure to bacteria and gluten are 2 of in the plasma, urine, or stool associated cussed in terms of anthropometric,
the more powerful triggers for zonulin- with the development of ASD in high-risk biochemical, and clinical parameters.74
induced tight junction disassembly. En- populations. Current literature that addresses
teric infections have been implicated in overall nutrient intakes in ASD does not
the pathogenesis of several pathologic Animal Models indicate a definitive consensus either
conditions, including allergic, autoim- The work in human subjects will be toward evidence for differing nutri-
mune, and inflammatory diseases, by enhanced by capitalizing on animal tional intake or similar nutritional in-
causing impairment of the intestinal models to investigate the mechanisms take in children with ASD compared
barrier. In addition to bacterial expo- of altered GI function and sensation in with typically developing children.
sure, gliadin, the environmental trigger ASDs. The outcome measures would There is also a body of studies targeted
of celiac disease, has been shown to include the histochemical and immu- specifically on intake and status of
alter the intestinal barrier function by nologic abnormalities seen in subsets nutrients related to bone health.68
releasing zonulin.61 Moreover, there is of ASD children, as well as the changes Problems in comparing existing stud-
increasing evidence supporting an in gut microbial composition observed ies include: (1) lack of adequate control
association of gut microbiota with in ASD. Specifically, animal models for groups in some studies; (2) variations
behavioral abnormalities such as ASD can be used to determine if there is in assessment tools and nutrient
anxiety and emotional reactivity51,62–65 a distinct GI phenotype in autism and if analysis programs; (3) different time
and potentially affecting 5-HT metab- such GI conditions lead to behaviors frames postdiagnosis; and (4) differ-
olism.66 When considered as a whole, associated with ASD. Animal models can ent reference values and “cutoffs”
there is much to learn regarding the also be used to examine potential defining inadequate. Table 1 summa-
integrity of the intestinal mucosa, its therapeutic options, for instance by rizes existing studies and significant
response to various gut microbiota, experimentally altering the gut micro- findings.
and the resultant effects on the body bial composition.53,67 Outcome meas-
systemically. Hediger et al45 reported that dairy-free
ures here would include behavior,
diets and unconventional food prefer-
immune status, neuropathology, as
Recommendations ences could place boys with autism
well as GI motility and sensitivity.
Current evidence indicates the impor- and ASDs at high risk for thinner, less
tance of better characterizing the un- dense bones (based on bone cortical
Nutrition
derlying pathology of GI problems in thickness measures) in comparison with
ASD and determining if there are any Current State of Knowledge a standardized reference based on
unique characteristics of GI dysfunction Several factors affect the nutritional age-matched typical boys. This occurred
specificto autism. Two recommendations status in children with ASDs, with even for those not on dairy-restricted
are made for research in this area. The most falling into 2 main categories: diets, although the differences were
first is to determine whether children (1) medical/nutritional factors and greater for those on casein-free diets.
with ASD differ from typically developing (2) behavioral factors.68–73 Medical/ Several factors have been implicated in
children in terms of the GI microbiome, nutritional factors encompass GI the higher risk for suboptimal bone
metabolites, inflammation, neurotrans- symptoms/problems, food allergies, development in children with ASDs, in-
mitters, immune response, and mucosal metabolic abnormalities, and/or pre- cluding lack of exercise, GI problems,
integrity. Second, it is important to de- existing nutrient deficiencies, as well and compromised vitamin D and cal-
termine if any confirmed differences in as nutrition-related medication side cium intake owing to either sensory/
these factors could be used to identify effects. Behavioral factors include texture issues, idiosyncratic eating
those with or at risk for developing ASD problem eating or idiosyncratic eat- patterns, and restricted diets, partic-
by using appropriate control groups. ing behaviors, sensory-processing ularly the gluten-free, casein-free
For example, 1 critical area would in- difficulties, and family factors.68 In diets.45,68,85 A recent nutrient intake
clude determining if mucosal 5HT this section, we focus on the medical study (based on 3-day food records) in

PEDIATRICS Volume 130, Supplement 2, November 2012 S163

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
TABLE 1 Studies Investigating Nutritional Quality in Children with Autism
Study No.of ASD No. of Control/ Dietary Tool Significant Findings
subjects Typical Subjects
Raiten and Massaro, 40 34 7-d Diet Record 1. ASD group had significantly greater intake of protein, carbohydrates,
198675 niacin, thiamin, riboflavin, calcium, phosphorus, and iron (P , .02)
2. No significant difference in vitamin A, vitamin C, or fat.
Ho et al, 199776 54 N/A 3-d Diet Record 1. Only 4 subjects with ASD (7.4%) met recommended servings from all
food groups.
2. All subjects had adequate protein intake, but had lower fat intake
and higher carbohydrate intake than recommended nutrient intake
for Canadians (RNI).
3. 42.6% of subjects were obese.
Cornish, 199877 17 N/A 3-d Dietary Recall 1. Nutrient intakes below RNI levels in 53% of children for one or more of
Food Frequency following: vitamin C, vitamin D, niacin, riboflavin, vitamin B6, calcium,
Questionnaire (FFQ) and zinc.
2. Calcium and niacin levels .200% or RNI in those children who drank
milk.
Lindsay et al, 200678 20 N/A FFQ 1. ,80% Recommended Dietary Allowance (RDA)/dietary reference
intake (DRI) considered inadequate
2. 45% consumed ↓calcium, 30% ↓pantothenic acid, 25% ↓ vitamin D,
40% ↓vitamin K
Levy et al, 200779 52 N/A 3-d Diet Record 1. The ASD subjects met 95%–101% of RDA guidelines for calories,
carbohydrates, and fat
2. ASD subjects overconsumed protein at 211% of RDA with a range of
67%–436% RDA among subjects
3. This study used 77% of RDA consumption as adequate diet
Lockner et al, 200880 20 20 3-d Diet Record 1. Vitamins E and A were the least likely to be met by Estimated Average
Requirement (EAR) for both groups
2. ASD subjects consumed less calcium and fiber, but with no established
EAR, significance was not determined
Schmitt et al, 200881 20 18 3-d Diet Record 1. This study defined adequate consumption as .67% of DRI
2. Both groups consumed ,67% for fiber
3. ASD group consumed ,67% for vitamins E and K
Johnson et al, 200871 19 15 24-h Recall 1. This study considered ,80% of RDAs or DRIs as inadequate
2. ASD group consumed significantly less vitamin K and significantly
fewer food choices from the vegetable group
Herndon et al, 200982 46 31 3-d Diet Record 1. ASD children consumed significantly less calcium but consumed
increasingly more vitamins B6 and E than controls.
2. ASD children consumed significantly more nondairy proteins and
fewer dairy items.
3. Both groups did not meet RDI for fiber, calcium, iron, vitamin D, and
vitamin E
Xia et al, 201083 111 N/A 3-d Diet Record 1. Comparison with Dietary Reference Intakes (DRI) for Chinese children.
2. Vitamin E and niacin exceeded 100% of DRI
3. Intakes were below DRI for vitamins A, B6 and C, Folic acid, calcium and
zinc
Bandini et al, 201071 53 58 Parent Interview 1. For both groups, fiber intake was inadequate as was intake of vitamin D,
vitamin E, and calcium
FFQ 2. For ASD children, inadequate intake of vitamin D and calcium was more
frequent
3-d Diet Diary 3. ASD children had a higher number of nutrients at inadequate intake
levels.
4. A limited food repertoire, but not food refusal, was associated with
higher nutritional inadequacy for both groups
Zimmer et al, 201184 22 22 FFQ 1. ASD children had higher intakes of magnesium
2. ASD children had lower intakes of protein, calcium, vitamin B12, and
vitamin D.
3. Selective eaters with ASD had greater likelihood of inadequate intake
of calcium, vitamin B12, zinc, and vitamin D
4. Selective eaters with ASD compared with nonselective eaters with ASD
had greater likelihood of inadequate intake of calcium.
5. ASD children had less food variety than typically developing children

S164 COURY et al
Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 SUPPLEMENT ARTICLE

children with ASDs ages 3 to 9 years on correcting identified deficiencies address the particular needs of an
revealed that the nutrients commonly and any additional benefits to the ASD population. Evidence-based algo-
analyzed as inadequate were those individual’s ASD symptoms. Further rithms specific for ASD should be de-
important for bone health (vitamins elucidation of the interrelationships veloped for each of the most common
A, D, and K) with 58.3%, 58.3%, and among these variables will assist in diagnoses that would address both
91.7% of the children consuming the establishment of clinical algo- management as well as treatment
intakes ,80% dietary reference in- rithms for categorization and effec- outcomes, tested rigorously, and sub-
take, respectively.86 tive treatment. sequently refined to optimize outcomes
(see this issue). These algorithms will
Collectively to date, these studies in-
provide the basis for comprehensive
dicate a trend for clinically significant TREATMENT AND OUTCOME guidelines and final recommendations
suboptimal nutrient intake in children
Current State of Knowledge provided to clinicians caring for chil-
with ASDs, with particular concern
dren with ASD.
related to bone health nutrients. In As highlighted in the previous sections,
general, nutritional status parameters children with ASDs frequently experi-
need to be refined and tailored for this ence GI symptoms, but their prevalence, Placebo-Controlled Trials
population. nature and, therefore, best treatments The positive impact of treatments aimed
remain elusive. Limited understanding at the GI problems in ASD children is
of the underlying pathology of GI con- widely reported by parents, but there
Recommendations
ditions in ASD limits the scientific ra- is little well-designed, controlled re-
Standardize Nutrient Assessment tionale for many of the interventions search to validate these interventions. It
Nutrient intake studies in ASDs should (eg, antifungal therapy, enzymes, and will also help substantially to stratify
be standardized by: (1) including con- nutritional supplements) aimed at the ASD population to identify specific
trol groups with typical children, (2) correcting these GI dysfunctions.9,10 A subgroups of children who may better
using consistent nutrient assessment possible theory unifying all the factors benefit from interventions. Many of
tools and analysis programs, (3) ac- mentioned above would link changes these approaches involve life-long
counting for the time frame post- in the gut microbiome with GI in- interventions (such as implementa-
diagnosis,and(4)establishing consistent flammation and other immunologic tion of a gluten-free diet), so better
reference values and acceptable cutoffs changes. identifying the individuals likely to re-
for defining inadequate intake. The most common GI diagnoses iden- spond to particular treatments can
tified in children with ASDs include reduce the costs and burden on families
Correlations With Nutritional Status constipation, diarrhea, and gastro- associated with nonefficacious treat-
esophageal reflux, and these are usu- ments. Stratification might be con-
Research should address the relation- ally treated in a standard manner.9,10 ducted by using clinical phenotypes or
ships between baseline nutritional Children with ASDs may not present through the identification and use of
status (by using standardized nutrient with the typical symptoms of a GI dis- specific biomarkers (see Table 2 ).
intake study principles and a thorough order, however, and an alteration of
nutrition assessment) and health status, The identification of specific ASD
their baseline behavior may be the microbiome and metabolic phenotypes
GI symptoms (constipation, diarrhea, only indicator of its existence. There is
flatulence, bloating, and nausea), bone a serious dearth of adequately designed
cortical thickness/bone mineral density, studies on treatments for documented TABLE 2 Biomarkers as Potential Outcome
behavior, sleep latency, food selectivity/ GI disorders and their outcomes, in- Measures
idiosyncratic behaviors, sensory- cluding behavioral changes, in children Biomarker Clinical Significance
processing difficulties (hypersensitiv- with ASDs. Intestinal Leaky gut
ity to certain food textures, tastes, or permeability
smells), and measures of inflammation Calprotectin Intestinal inflammation
(eg, cytokines, c-reactive protein). Recommendations Celiac disease Celiac disease and gluten
serology tests sensitivity
Treatment Guidance Food allergy panel Food allergies
Organic acid testing Vitamin B12 or folic acid
Efficacy of Nutritional Intervention Although general pediatric guidelines deficiencies
Studies should determine the effec- exist for specific GI conditions, the Analysis gut Gut dysbiosis
tiveness of nutritional interventions recommendations may not always microbiota

PEDIATRICS Volume 130, Supplement 2, November 2012 S165

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
TABLE 3 Key Takeaway Messages microbiota in ASD, but also on other
Epidemiology metabolic imbalances that may be
• Gastrointestinal conditions in ASD are at least as common as in typically developing population present. This knowledge, in turn, could
• It is less clear whether GI conditions are more common in children with ASD aid in the development of strategies for
• Individuals with ASDs should receive diagnostic evaluations and medical treatments for GI disorders just as
children without ASD do
novel approaches for the prevention/
Biology treatment of ASD.
• Several studies identify abnormalities of the immune system in ASD
• Some GI conditions in ASD may have an immunologic basis
• Some individuals with ASD have abnormalities in serum concentrations of the neurotransmitter serotonin DISCUSSION
• Alterations in serotonin may be associated with anxiety, GI symptoms, and bone density abnormalities
Nutrition Clinical experience, parent concern, and
• Several studies have identified abnormalities of diet and altered bone mineral density, but the cause developing lines of research all make
(dietary preference, allergic issues, or absorptive problems) remains uncertain.
Treatment clear the importance of addressing GI
• No treatments for GI problems specific to children with ASD currently exist. problems and symptoms among chil-
• Individuals with ASD and GI conditions appear to respond to treatment much as people with typical dren and adolescents with ASD. Although
development do.
• Studies that can lead to management that addresses specific challenges of the disorder are still needed. thissymposium and reviewindicatehuge
gaps in evidence, especially to support
interventions to improve the health and
TABLE 4 Key Research Objectives for an Integrated Approach to Addressing Knowledge Gaps GI symptoms in ASD, several promising
1. Determine the pathology of gastrointestinal conditions in ASDs through integrated studies of the gut areas of research bring optimism and
microbiome, metabolism, inflammation, immunity, and mucosal integrity. The outcome measures would focus to the key next steps in in-
include the histochemical and immunologic abnormalities seen in subsets of ASD children, as well as the vestigation and treatment trials. Table 3
changes in gut microbial composition observed in ASDs.
2. Increase the use of animal models to better understand the underlying pathology. Studies manipulating the indicates key messages from this work.
microbiome in animal models may help promote better understanding of these conditions. Subsequent to this symposium, others
3. Identify biomarkers for assessing the status of these conditions and to guide identification and treatment. endorsed many of the same points.9,10
4. Identify biomarkers of nutritional status that will not only guide monitoring of response to treatment but
help identify those requiring intervention.
This article reviews key areas of epi-
5. Identify behavioral phenotypes related to poor nutritional status to better identify and treat this population. demiology, underlying pathology, nu-
6. Develop evidence-based clinical algorithms to help guide clinicians in the evaluation and treatment of trition, and treatment and outcome.
gastrointestinal problems in individuals with ASDs.
Ongoing efforts should also address
how to integrate diverse areas of re-
can also help to define additional di- microbiota composition. These findings search and findings into better un-
agnostic tools, biomarkers, and thera- may have a far-reaching impact not derstanding of the complex activities
peutic interventions such as the use only on our understanding of the role of the GI system in children with ASDs
of probiotics to change the intestinal of specific allergens, leaky gut, and (Table 4).

REFERENCES
1. Centers for Disease Control and Pre- 5. Erickson CA, Stigler KA, Corkins MR, Posey 9. Buie T, Campbell DB, Fuchs GJ III, et al.
vention. Prevalence of autism spectrum DJ, Fitzgerald JF, McDougle CJ. Gastroin- Evaluation, diagnosis, and treatment of
disorders: Autism and Developmental Dis- testinal factors in autistic disorder: a criti- gastrointestinal disorders in individuals
abilities Monitoring Network, United States, cal review. J Autism Dev Disord. 2005;35(6): with ASDs: a consensus report. Pediatrics.
2006. MMWR Surveill Summ. 2009;58(SS- 713–727 2010;125(suppl 1):S1–S18
10):1–20 10. Buie T, Fuchs GJ III, Furuta GT, et al. Rec-
6. Fombonne E, Cook EH. MMR and autistic
2. Kogan MD, Blumberg SJ, Schieve LA, et al. enterocolitis: consistent epidemiological ommendations for evaluation and treat-
Prevalence of parent-reported diagnosis of ment of common gastrointestinal problems
failure to find an association. Mol Psychi-
autism spectrum disorder among children in in children with ASDs. Pediatrics. 2010;125
atry. 2003;8(2):133–134
the US, 2007. Pediatrics. 2009;124(5):1395–1403 (suppl 1):S19–S29
7. DeStefano F. MMR vaccine and autism:
3. Horvath K, Perman JA. Autism and gastro- 11. Black C, Kaye JA, Jick H. Relation of child-
a review of the evidence for a causal as-
intestinal symptoms. Curr Gastroenterol hood gastrointestinal disorders to autism:
Rep. 2002;4(3):251–258 sociation. Mol Psychiatry. 2002;7(suppl 2): nested case-control study using data from
4. Filipek PA, Accardo PJ, Baranek GT, et al. S51–S52 the UK General Practice Research Data-
The screening and diagnosis of autistic 8. Galiatsatos P, Gologan A, Lamoureux E. Au- base. BMJ. 2002;325(7361):419–421
spectrum disorders. J Autism Dev Disord. tistic enterocolitis: fact or fiction? Can J 12. Fombonne E, Chakrabarti S. No evidence
1999;29(6):439–484 Gastroenterol. 2009;23(2):95–98 for a new variant of measles-mumps-rubella-

S166 COURY et al
Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 SUPPLEMENT ARTICLE

induced autism. Pediatrics. 2001;108(4). 24. Onore C, Careaga M, Ashwood P. The role of 37. Coates MD, Mahoney CR, Linden DR, et al.
Available at: www.pediatrics.org/cgi/content/ immune dysfunction in the pathophysiology Molecular defects in mucosal serotonin
full/108/4/e58 of autism. Brain Behav Immun. 2012;26(3): content and decreased serotonin reuptake
13. Xue Ming , Brimacombe M, Chaaban J, 383–392 transporter in ulcerative colitis and irrita-
Zimmerman-Bier B, Wagner GC. Autism 25. Ashwood P, Enstrom A, Van de Water J. ble bowel syndrome. Gastroenterology.
spectrum disorders: concurrent clinical dis- Autism, gastrointestinal disturbance, and 2004;126(7):1657–1664
orders. J Child Neurol. 2008;23(1):6–13 immune dysfunction: what is the link? In: 38. Magro F, Vieira-Coelho MA, Fraga S, et al.
14. Molloy CA, Manning-Courtney P. Prevalence Chauhan A, Chauhan V, Brown TW, eds. Impaired synthesis or cellular storage
of chronic gastrointestinal symptoms in Autism: Oxidative Stress, Inflammation, and of norepinephrine, dopamine, and 5-
children with autism and autistic spectrum Immune Abnormalities. Boca Raton, FL: CRC hydroxytryptamine in human inflammatory
disorders. Autism. 2003;7(2):165–171 Press; 2010:278–298 bowel disease. Dig Dis Sci. 2002;47(1):216–
15. Nikolov RN, Bearss KE, Lettinga J, et al. 26. Atladóttir HO, Pedersen MG, Thorsen P, et al. 224
Gastrointestinal symptoms in a sample of Association of family history of autoim- 39. Ahonen A, Kyösola K, Penttilä O. Entero-
children with pervasive developmental mune diseases and autism spectrum dis- chromaffin cells in macrophages in ulcer-
disorders. J Autism Dev Disord. 2009;39(3): orders. Pediatrics. 2009;124(2):687–694 ative colitis and irritable colon. Ann Clin
405–413 27. Comi AM, Zimmerman AW, Frye VH, Law PA, Res. 1976;8(1):1–7
16. Parracho HMRT, Bingham MO, Gibson GR, Peeden JN. Familial clustering of autoim- 40. El-Salhy M, Danielsson A, Stenling R,
McCartney AL. Differences between the gut mune disorders and evaluation of medical Grimelius L. Colonic endocrine cells in in-
microflora of children with autistic spec- risk factors in autism. J Child Neurol. 1999; flammatory bowel disease. J Intern Med.
trum disorders and that of healthy chil- 14(6):388–394 1997;242(5):413–419
dren. J Med Microbiol. 2005;54(pt 10): 28. Enstrom AM, Van de Water JA, Ashwood P. 41. Linden DR, Chen JX, Gershon MD, Sharkey
987–991 Autoimmunity in autism. Curr Opin Investig KA, Mawe GM. Serotonin availability is in-
17. Smith RA, Farnworth H, Wright B, Allgar V. Drugs. 2009;10(5):463–473 creased in mucosa of guinea pigs with
Are there more bowel symptoms in chil- TNBS-induced colitis. Am J Physiol Gastro-
29. Cabanlit M, Wills S, Goines P, Ashwood P,
dren with autism compared to normal intest Liver Physiol. 2003;285(1):G207–G216
Van de Water J. Brain-specific autoanti-
children and children with other de- 42. Costedio MM, Hyman N, Mawe GM. Seroto-
bodies in the plasma of subjects with au-
velopmental and neurological disorders?: A nin and its role in colonic function and in
tistic spectrum disorder. Ann N Y Acad Sci.
case control study. Autism. 2009;13(4): gastrointestinal disorders. Dis Colon Rec-
2007;1107:92–103
343–355 tum. 2007;50(3):376–388
30. Croen LA, Braunschweig D, Haapanen L,
18. Taylor B, Miller E, Lingam R, Andrews N, 43. Bischoff SC, Mailer R, Pabst O, et al. Role
et al. Maternal mid-pregnancy autoanti-
Simmons A, Stowe J. Measles, mumps, and of serotonin in intestinal inflammation:
bodies to fetal brain protein: the early
rubella vaccination and bowel problems or knockout of serotonin reuptake trans-
markers for autism study. Biol Psychiatry.
developmental regression in children with porter exacerbates 2,4,6-trinitrobenzene
2008;64(7):583–588
autism: population study. BMJ. 2002;324 sulfonic acid colitis in mice. Am J Physiol
(7334):393–396 31. Singer HS, Morris CM, Gause CD, Gillin PK, Gastrointest Liver Physiol. 2009;296(3):
Crawford S, Zimmerman AW. Antibodies G685–G695
19. Valicenti-McDermott M, McVicar K, Rapin I,
against fetal brain in sera of mothers with
Wershil BK, Cohen H, Shinnar S. Frequency 44. Ghia JE, Li N, Wang H, et al. Serotonin has
autistic children. J Neuroimmunol. 2008; a key role in pathogenesis of experimental
of gastrointestinal symptoms in children
194(1-2):165–172 colitis. Gastroenterology. 2009;137(5):
with autistic spectrum disorders and as-
sociation with family history of autoim- 32. Zimmerman AW, Connors SL, Matteson KJ, 1649–1660
mune disease. J Dev Behav Pediatr. 2006;27 et al. Maternal antibrain antibodies in 45. Hediger ML, England LJ, Molloy CA, Yu KF,
(suppl 2):S128–S136 autism. Brain Behav Immun. 2007;21(3): Manning-Courtney P, Mills JL. Reduced
20. Wang LW, Tancredi DJ, Thomas DW. The 351–357 bone cortical thickness in boys with autism
prevalence of gastrointestinal problems in 33. Wills S, Cabanlit M, Bennett J, Ashwood P, or autism spectrum disorder. J Autism Dev
children across the United States with au- Amaral DG, Van de Water J. Detection of Disord. 2008;38(5):848–856
tism spectrum disorders from families autoantibodies to neural cells of the cere- 46. Patterson PH. Modeling autistic features
with multiple affected members. J Dev bellum in the plasma of subjects with au- in animals. Pediatr Res. 2011;69(5 pt 2):
Behav Pediatr. 2011;32(5):351–360 tism spectrum disorders. Brain Behav 34R–40R
21. Ibrahim SH, Voigt RG, Katusic SK, Weaver AL, Immun. 2009;23(1):64–74 47. Martin LA, Ashwood P, Braunschweig D,
Barbaresi WJ. Incidence of gastrointestinal 34. Torrente F, Anthony A, Heuschkel RB, Cabanlit M, Van de Water J, Amaral DG.
symptoms in children with autism: a pop- Thomson MA, Ashwood P, Murch SH. Focal- Stereotypies and hyperactivity in rhesus
ulation-based study. Pediatrics. 2009;124 enhanced gastritis in regressive autism monkeys exposed to IgG from mothers of
(2):680–686 with features distinct from Crohn’s and children with autism. Brain Behav Immun.
22. Afzal N, Murch S, Thirrupathy K, Berger L, Helicobacter pylori gastritis. Am J Gastro- 2008;22(6):806–816
Fagbemi A, Heuschkel R. Constipation with enterol. 2004;99(4):598–605 48. Bauman MD, Iosif AM, Smith SEP, et al. A
acquired megarectum in children with au- 35. Mayer EA. Gut feelings: the emerging bi- nonhuman primate model of maternal im-
tism. Pediatrics. 2003;112(4):939–942 ology of gut-brain communication. Nat Rev mune activation. SFN Abstracts. 2011;
23. Careaga M, Van de Water J, Ashwood P. Neurosci. 2011;12(8):453–466 778.06/Y29
Immune dysfunction in autism: a pathway 36. Bercik P, Collins SM, Verdu EF. Microbes 49. Malkova NV, Yu CZ, Hsiao EY, Moore MJ,
to treatment. Neurotherapeutics. 2010;7(3): and the gut-brain axis. Neurogastroenterol Patterson PH. Maternal immune activation
283–292 Motil. 2012;24(5):405–413 yields offspring displaying mouse versions

PEDIATRICS Volume 130, Supplement 2, November 2012 S167

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 of the three core symptoms of autism. prehaptoglobin-2. Proc Natl Acad Sci USA. therapies in autism: a spectrum of what we
Brain Behav Immun. 2012;26(4):607–616 2009;106(39):16799–16804 know: Part 2. Infant Child Adolesc Nutr.
50. Bercik P, Park AJ, Sinclair D, et al. The an- 62. Heijtz RD, Wang S, Anuar F, et al. Normal gut 2010;2(2):120–133
xiolytic effect of Bifidobacterium longum microbiota modulates brain development 75. Raiten DJ, Massaro T. Perspectives on the
NCC3001 involves vagal pathways for gut- and behavior. Proc Natl Acad Sci USA. 2011; nutritional ecology of autistic children.
brain communication. Neurogastroenterol 108(7):3047–3052 J Autism Dev Disord. 1986;16(2):133–143
Motil. 2011;23(12):1132–1139 63. Forsythe P, Sudo N, Dinan T, Taylor VH, 76. Ho H, Eaves LC, Peabody D. Nutrient intake
51. Bravo JA, Forsythe P, Chew MV, et al. In- Bienenstock J. Mood and gut feelings. Brain and obesity in children with autism. Focus
gestion of Lactobacillus strain regulates Behav Immun. 2010;24(1):9–16 Autism Other Dev Disabl. 1997;12(3):187–
emotional behavior and central GABA re- 64. Cryan JF, O’Mahony SM. The microbiome- 192
ceptor expression in a mouse via the vagus gut-brain axis: from bowel to behavior. 77. Cornish E. A balanced approach toward
nerve. Proc Natl Acad Sci USA. 2011;108 Neurogastroenterol Motil. 2011;23(3):187– healthy eating in autism. J Hum Nutr Diet.
(38):16050–16055 192 1998;11(6):501–509
52. Li W, Dowd SE, Scurlock B, Acosta-Martinez 65. Collins SM, Bercik P. The relationship be- 78. Lindsay RL, Eugene Arnold L, Aman MG,
V, Lyte M. Memory and learning behavior in tween intestinal microbiota and the central et al. Dietary status and impact of risper-
mice is temporally associated with diet- nervous system in normal gastrointestinal idone on nutritional balance in children
induced alterations in gut bacteria. Phys- function and disease. Gastroenterology. with autism: a pilot study. J Intellect Dev
iol Behav. 2009;96(4-5):557–567 2009;136(6):2003–2014 Disabil. 2006;31(4):204–209
53. Lyte M. Probiotics function mechanistically 66. Desbonnet L, Garrett L, Clarke G, Bienenstock 79. Levy S, Souders MC, Ittenbach RF, Giarelli E,
as delivery vehicles for neuroactive com- J, Dinan TG. The probiotic Bifidobacteria Mulberg A, Pinto-Martin J. Relationship of
pounds: Microbial endocrinology in the infantis: An assessment of potential anti- dietary intake to gastrointestinal symp-
design and use of probiotics. Bioessays. depressant properties in the rat. J Psy- toms in children with autism spectrum
2011;33(8):574–581 chiatr Res. 2008;43(2):164–174 disorders. Biol Psychol. 2007;61(4):492–497
54. D’Eufemia P, Celli M, Finocchiaro R, et al. 67. Round JL, Mazmanian SK. The gut micro- 80. Lockner DW, Crowe TK, Skipper BJ. Dietary
Abnormal intestinal permeability in chil- biota shapes intestinal immune responses intake and parents’ perception of mealtime
dren with autism. Acta Paediatr. 1996;85(9): during health and disease. Nat Rev Immu- behaviors in preschool-age children with
1076–1079 nol. 2009;9(5):313–323 autism spectrum disorder and in typically
55. de Magistris L, Familiari V, Pascotto A, et al. 68. Geraghty ME, Depasquale GM, Lane AE. Nu- developing children. J Am Diet Assoc. 2008;
Alterations of the intestinal barrier in patients tritional intake and therapies in autism: 108(8):1360–1363
with autism spectrum disorders and in a spectrum of what we know: Part 1. Infant 81. Schmitt L, Heiss CJ, Campbell E. A com-
their first-degree relatives. J Pediatr Gas- Child Adolesc Nutr. 2010;2:62–69 parison of nutrient intake and eating
troenterol Nutr. 2010;51(4):418–424 69. Ahearn WH, Castine T, Nault K, Green G. An behaviors of boys with and without autism.
56. Altieri L, Neri C, Sacco R, et al. Urinary p- assessment of food acceptance in children Topics Clin Nutr. 2008;23(1):23–31
cresol is elevated in small children with with autism or pervasive developmental 82. Herndon AC, DiGuiseppi C, Johnson SL,
severe autism spectrum disorder. Bio- disorder-not otherwise specified. J Autism Leiferman J, Reynolds A. Does nutritional
markers. 2011;16(3):252–260 Dev Disord. 2001;31(5):505–511 intake differ between children with autism
57. Finegold SM. Therapy and epidemiology of 70. Bandini LG, Anderson SE, Curtin C, et al. spectrum disorders and children with
autism-clostridial spores as key elements. Food selectivity in children with autism typical development? J Autism Dev Disord.
Med Hypotheses. 2008;70:508–511 spectrum disorders and typically developing 2009;39(2):212–222
58. Finegold SM, Dowd SE, Gontcharova V, et al. children. J Pediatr. 2010;157(2):259–264 83. Xia W, Zhou Y, Sun C, Wang J, Wu L. A pre-
Pyrosequencing study of fecal microflora of 71. Johnson CR, Handen BL, Mayer-Costa M, liminary study on nutritional status and
autistic and control children. Anaerobe. Sacco K. Eating habits and dietary status in intake in Chinese children with autism. Eur
2010;16(4):444–453 young children with autism. J Dev Phys J Pediatr. 2010;169(10):1201–1206
59. Liu Z, Li N, Neu J. Tight junctions, leaky Disabil. 2008;20:437–448 84. Zimmer M, Hart L, Manning-Courtney P,
intestines, and pediatric diseases. Acta 72. Schreck KA, Williams K. Food preferences Murray D, Bing N, Summer S. Food variety
Paediatr. 2005;94(4):386–393 and factors influencing food selectivity for as a predictor of nutritional status among
60. Fasano A. Physiological, pathological, and children with autism spectrum disorders. children with autism. J Autism Dev Disord.
therapeutic implications of zonulin-mediated Res Dev Disabil. 2006;27(4):353–363 2012;42(4):549–556
intestinal barrier modulation: living life on 73. Williams K, Gibbons BG, Schreck KA. Com- 85. Holick MF. Vitamin D deficiency. N Engl J
the edge of the wall. Am J Pathol. 2008;173 paring selective eaters with and without Med. 2007;357(3):266–281
(5):1243–1252 developmental disabilities. J Dev Phys Dis- 86. Altenburger J, Geraghty ME, Wolf K, Taylor
61. Tripathi A, Lammers KM, Goldblum S, et al. abil. 2005;17(3):299–309 CA, Lane AE. The quality of nutritional intake
Identification of human zonulin, a physio- 74. Geraghty ME, Bates-Wall J, Ratliff-Schaub K, in children with autism. J Am Diet Assoc.
logical modulator of tight junctions, as Lane AE. Nutritional interventions and 2010;110(9):A40

S168 COURY et al
Downloaded from pediatrics.aappublications.org by guest on February 22, 2014
 Gastrointestinal Conditions in Children With Autism Spectrum Disorder:
Developing a Research Agenda
Daniel L. Coury, Paul Ashwood, Alessio Fasano, George Fuchs, Maureen Geraghty,
Ajay Kaul, Gary Mawe, Paul Patterson and Nancy E. Jones
Pediatrics 2012;130;S160
DOI: 10.1542/peds.2012-0900N
Updated Information & including high resolution figures, can be found at:
Services http://pediatrics.aappublications.org/content/130/Supplement
_2/S160.full.html
References This article cites 82 articles, 20 of which can be accessed free
at:
http://pediatrics.aappublications.org/content/130/Supplement
_2/S160.full.html#ref-list-1
Citations This article has been cited by 2 HighWire-hosted articles:
http://pediatrics.aappublications.org/content/130/Supplement
_2/S160.full.html#related-urls
Subspecialty Collections This article, along with others on similar topics, appears in
the following collection(s):
Development/Behavioral Issues
http://pediatrics.aappublications.org/cgi/collection/developme
nt:behavioral_issues_sub
Autism/ASD
http://pediatrics.aappublications.org/cgi/collection/autism:asd
_sub
Gastroenterology
http://pediatrics.aappublications.org/cgi/collection/gastroente
rology_sub
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
http://pediatrics.aappublications.org/site/misc/Permissions.xh
tml
Reprints Information about ordering reprints can be found online:
http://pediatrics.aappublications.org/site/misc/reprints.xhtml

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright © 2012 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org by guest on February 22, 2014